General Hardware Discussion - Page 2

jvang · 2020-10-06, 00:27

Quote:

Originally Posted by Nick

You're right, it's confusing! That's because what it gets used for has changed so quickly.
This Intel document is a few years old now but still gives a good introduction:
http://www.intel.com/content/dam/www...149x-paper.pdf

I've mainly seen it used for debugging embedded systems
(processors in non-computers - anything from a settop box for satellite TV to a washing machine).
But, to start with, you just need to understand enough to work out whether this is a useful interface for you
to get deeper into the hardware, or whether there is something else more suitable available.

Quote:

Originally Posted by LaurV

Almost all devices that include a not-the-lowest-performance processor, say they have at least one 32-bit or 16-bit "brain" (like CPU, MCU, chipset, FPGA, all kind of controllers, etc) have a JTAG or similar (SWD, SWIM), interface. But unless the documentation is provided, nobody know how to use it, except the designers/manufacturers.

JTAG per se, is a very simple thing, just a VERY-VERY big serial shift register (google: shift register). Imagine all bits inside of your CPU, all registers, all internal settings, input and output pin values, etc., placed in a very-very long string/queue. You can clock them out bit by bit and read them, or you can clock them in bit by bit and set them into the CPU, just by using three lines, one "data input", one "data output" and one "clock" line. Every pulse you give on the clock line will shift in a bit on DI, and shift out a bit on DO. You can shift in a number of bits till they "align" with internal registers, then set them all in the same time. Or read all internal registers at once, them shift them out one by one and see what they contain. That's all.

But the alignment, you have no way to guess, until the manufacturer provides you with the datasheet.

We use JTAG interfaces in our daily work, and could not do half of out job without them. They are used to program and debug almost all the hardware we make. Very few we make public.

We don't think jvang needs JTAG access to his toys, unless he kills the BIOS and needs flash rewrite by external artillery (programmer/debugger, etc). The BIOS of the mobo itself, and high-level utilities (i.e. run from OS) to play with that BIOS, should provide all the tools needed to do what he wants.

how do you even use it? is there a device that you plug into the hardware for this purpose?

thought that samsung's new 980 pro coming out was a weird drive, as theyve dropped the 2 bit mlc for tlc, so aside from burst speeds while the SLC cache persists, its much worse than the old 970 pro. lesser endurance, lesser sustained speeds, that sorta thing. i think its a bit weird how endurance ratings on ssds are still kinda low, as i saw some samsung pm1725a's (3.2tb) on sale the other day which advertise a 29 petabyte endurance rating! i get that enterprise anything is gonna be really overengineered but thats kind of a lot. uses a x8 pcie 3.0 interface as its a bit of an older drive, 6gbps read/2gbps write. was almost tempted to get one, but they were 500 bucks apiece used lol

LaurV · 2020-10-06, 03:14

Quote:

Originally Posted by jvang

how do you even use it? is there a device that you plug into the hardware for this purpose?

You use a JTAG adapter, which is usually composed from a small cable (3 to 8 wires, the standard connectors are either 6 pins or 20 pins, but most of them not used for this purpose) and a small PCBA which you can buy for a couple of bucks or build yourself, nowadays the most common is "JTAG to RS232" or "JTAG to USB". The boards that you want to "hack" will have always a 3, 4, 5, or 6 pins/pads on the PCB, close to the IC (there may be one such "connector" for every controller/IC on the board). You connect the JTAG adaptor between the board you want to debug/hack and your computer using a normal USB cable. The hardware itself is piece of cake. The rest is software. There are software tools (like the one from Lauterbach (google it) we used to debug our PXA270 and PXA320 boards, years ago) which cost tens of thousands of dollars.

To come with an analogy: Imagine you go inside of the Intel building with a spool of very long string, and go into ALL offices, and tie the string around the hand of every person you meet. From general manager, engineers, accountants, to janitors. Then you go out from the building, having the end of the string in your hand. That's JTAG. The string is very flexible, it will not impede anybody to do his/her job, in fact, they will have no idea that there is a string attached to their hand. They still can talk to each-other, and hold their meetings, etc. But if you pull the string once, then everybody, and I mean EVERYBODY in the building will write on a piece of paper the word they were saying or thinking in EXACTLY that moment, and attach the piece of paper to the string. Every subsequent pulls you do, everybody will take the piece of paper in front of him and move it in front of his next colleague on the string. The last guy from the string will give the piece of paper to you. You also can give a piece of paper with whatever word you want, to the first guy on the string, who now has no paper. You can repeat this process as many times you want, until you get all the words out, or until you get bored. At the end, you will have one word from every person in the building, and they each will have one new word in front of him/her. Then you say, "ok, I finished", or "I am bored", and stop pulling the string. In this moment, all of them continue their work, but will say/use/think the new words instead of their original words.

Yeah, I know, a big mess, haha

, you can make the accountant say to the general manager "fluck you" instead of "yes sir". You can make any trouble you want, including making the janitor set the building on fire. This is not a metaphor, beside of the fact that the building is here a substitute for the IC, and the employees are the internal registers, but yes, you CAN set the IC on fire (for example, using some outputs two by two to short circuit Vcc to Ground, as the output values are inside internal registers).

This is JTAG. Repeat as many times you want.

Every chip manufacturer that offers a chip that supports JTAG will implement this "shift register", as it is called, in the respective chip (it HAS additional hardware). Most chips nowadays support JTAG. The big advantage of JTAG is that it is very simple to implement, doesn't need much silicon, it doesn't influence the operation of the IC, and most important, it doesn't require extension of the package (which is the most expensive, silicon is not expensive, but going from a 64 pin IC to a 100 pin IC will almost double its manufacturing price). Imagine holding in your hand not only one wire, but 573 wires, each attached to one employee...

Also, the external tools are cheaper (as I said, you can use a 3 dollars board, connected to few wires, instead of having a few hundreds of wires cables, connected to a desk-sized debugger which costs thousands of dollars - that was the standard before JTAG, you can google ICE/ICD - in-circuit emulators/debuggers, they were HUGE machines, taking a lot of space on your desk, and costing "millions", not counting the fact that, to debug a board with 5 controllers on it in real time, you would need 5 such "tanks" on your desk. Nowadays, as a designer, you can connect all controllers on board in a single JTAG chain if you want, or populate separate connectors for each.

But then, imagine I come to the Intel building above, and kick your butt and take the string out of your hand and I start pulling out words. They will mean nothing for me, because I have no idea of the order on which you tied the people in row. After a while I can deduct that the person 137 is the chief accountant, because he is using too often words like bank and invoice, or that the person 351 is the janitor as he talks about cleaning the floors, but without a complete documentation from you, I will not be able to do much with the string and the words, beside of creating an even bigger mess..

So, how do you use it? the first step is to take the datasheet and the reference manual of the chip you want to debug/hack, and read the chapters about JTAG and debugging...

Nick · 2020-10-06, 07:59

OK well LaurV clearly doesn't think that that will be a quick way to get a deeper insight into the hardware, and he's an expert at this stuff!
If you like hardware enough to consider reading a serious book on it, then the place to start is with the books of John Hennessy and David Patterson:
"Computer Architecture" or "Computer Organization and Design".
Your Dad says you like watching youtube: they gave a talk when they received the Turing award a couple of years back, and you can see that here:
https://www.acm.org/hennessy-patterson-turing-lecture

Xyzzy · 2020-10-06, 22:47

LaurV: Do you know of an easy way to crimp fan connectors without this super expensive tool?

We have a pile of fans that need to have new tails put on them. (Finding the connectors and little crimp tabs isn't easy either.)

See: http://tech.mattmillman.com/info/crimpconnectors/#kk100

Maybe the "PA-09" is okay?

Uncwilly · 2020-10-07, 00:01

Why not buy 'adapters' or other similar leads (or find donor parts), cut and splice them. Use a bit of heat shrinkable tubing to protect from shorts.

LaurV · 2020-10-07, 05:43

Quote:

Originally Posted by Xyzzy

LaurV: Do you know of an easy way to crimp fan connectors without this super expensive tool?

We have the tool (actually, the factory we work for, has the tool). But for our private needs, when we can't bring the wires with us to the factory (which is most of the times, when we succeed to break some wires at home, we are always in hurry to fix them and can't wait till next day) a pair of pliers and the soldering iron is the way to go. With a bit of skill you can bend the crimping contacts nice, and anyhow, tool or not tool, I won't trust them without a bit of soldered tin on top..

One other way is to have a stock of crimped wires (crimp a bunch of them one day, and take them home, or just rip them off from other connectors when needed) and "do the surgery" if the crimp is broken. Surgery involves cutting the broken wire, soldering a new wire, insulate with a bit of heat-shrink tube. Here some of my (actually, JTAG/SWIM (!)) cables with such surgery done.

If you talk "industrial" and "mass production", buy the crimping tool. It worth.

jvang · 2020-10-09, 22:17

Quote:

Originally Posted by Uncwilly

Why not buy 'adapters' or other similar leads (or find donor parts), cut and splice them. Use a bit of heat shrinkable tubing to protect from shorts.

actually, thats been done on the fans i have and the spliced section is very stiff and hard to maneuver around for managing the cable. not ideal but it is an option, though premade extensions are a bit pricey.

Quote:

Originally Posted by LaurV

We have the tool (actually, the factory we work for, has the tool). But for our private needs, when we can't bring the wires with us to the factory (which is most of the times, when we succeed to break some wires at home, we are always in hurry to fix them and can't wait till next day) a pair of pliers and the soldering iron is the way to go. With a bit of skill you can bend the crimping contacts nice, and anyhow, tool or not tool, I won't trust them without a bit of soldered tin on top..

One other way is to have a stock of crimped wires (crimp a bunch of them one day, and take them home, or just rip them off from other connectors when needed) and "do the surgery" if the crimp is broken. Surgery involves cutting the broken wire, soldering a new wire, insulate with a bit of heat-shrink tube. Here some of my (actually, JTAG/SWIM (!)) cables with such surgery done.

If you talk "industrial" and "mass production", buy the crimping tool. It worth.

been doing a lot of research on cable sleeving/crimping and all that, seems pretty complicated! i'll start out with the few cables i've got, just need to get replacement pins for the connectors to begin with.

so i was thinking about gpus and the new ampere cards, and the performance per watt seems strange. in like every recent generation there's been a pretty decent upgrade in efficiency compared to the last; for instance, comparing the 1070 to the 980ti, or the 2080ti versus the 1080ti. in the first case, you have the same or better performance for maybe half the wattage (my 980ti can use up to 360 watts easily....), and in the second case the 2080ti uses similar or slightly less power for more performance. but comparing ampere to turing, the 3080 uses 20-30% more power than the 2080ti in a given workload for 25-35% more performance, on average, which just seems a bit weird in comparison to past generations. if the 3070 matches the 2080ti in performance as advertised (unlikely lol), its still rated for a 220w tdp which, while making it a good bit more efficient than the 3080 or 2080ti, doesnt seem spectacular compared to previous generations. any thoughts on all that?

LaurV · 2020-10-13, 15:27

Hihi, I expected a comment on the fact that on the manometer, the green area is from 31 to 61, and the yellow is from 61 to 89

(and there are also other prime-related tricks on that screen, but maybe the image is not very clear).

jvang · 2020-10-13, 19:24

Quote:

Originally Posted by LaurV

Hihi, I expected a comment on the fact that on the manometer, the green area is from 31 to 61, and the yellow is from 61 to 89

(and there are also other prime-related tricks on that screen, but maybe the image is not very clear).

A manometer is for measuring pressure, no? Hard to read it anyways lol

Slotted in that 3900x the other day, kept my settings from the 3600 so its just running with stock settings, no tdp limits, and 2x16gb ram at 3600mhz 14-15-15-30 1.45v. Unsurprisingly, it's just shy of two times as fast in multithreaded workloads, who wouldve thought...

Very lengthy, but here are some p95 times for the 3600:

Code:

Prime95 64-bit version 30.3, RdtscTiming=1
Timings for 560K FFT length (1 core, 1 worker):  1.74 ms.  Throughput: 574.95 iter/sec.
Timings for 560K FFT length (1 core hyperthreaded, 1 worker):  1.63 ms.  Throughput: 615.14 iter/sec.
Timings for 560K FFT length (2 cores, 1 worker):  0.92 ms.  Throughput: 1090.08 iter/sec.
Timings for 560K FFT length (2 cores, 2 workers):  1.75,  1.73 ms.  Throughput: 1150.49 iter/sec.
Timings for 560K FFT length (2 cores hyperthreaded, 1 worker):  0.86 ms.  Throughput: 1168.70 iter/sec.
Timings for 560K FFT length (2 cores hyperthreaded, 2 workers):  1.65,  1.66 ms.  Throughput: 1207.65 iter/sec.
Timings for 560K FFT length (3 cores, 1 worker):  0.67 ms.  Throughput: 1500.02 iter/sec.
Timings for 560K FFT length (3 cores, 2 workers):  0.92,  1.74 ms.  Throughput: 1660.88 iter/sec.
Timings for 560K FFT length (3 cores, 3 workers):  1.75,  1.74,  1.76 ms.  Throughput: 1715.52 iter/sec.
Timings for 560K FFT length (3 cores hyperthreaded, 1 worker):  0.62 ms.  Throughput: 1620.17 iter/sec.
Timings for 560K FFT length (3 cores hyperthreaded, 2 workers):  0.86,  1.72 ms.  Throughput: 1747.95 iter/sec.
Timings for 560K FFT length (3 cores hyperthreaded, 3 workers):  1.67,  1.67,  1.66 ms.  Throughput: 1803.18 iter/sec.
Timings for 560K FFT length (4 cores, 1 worker):  0.59 ms.  Throughput: 1691.11 iter/sec.
Timings for 560K FFT length (4 cores, 2 workers):  0.92,  0.94 ms.  Throughput: 2154.84 iter/sec.
Timings for 560K FFT length (4 cores, 3 workers):  1.76,  1.77,  0.93 ms.  Throughput: 2205.22 iter/sec.
Timings for 560K FFT length (4 cores, 4 workers):  1.79,  1.76,  1.80,  1.74 ms.  Throughput: 2254.46 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 1 worker):  0.52 ms.  Throughput: 1913.90 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 2 workers):  0.88,  0.87 ms.  Throughput: 2285.68 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 3 workers):  1.77,  1.68,  0.88 ms.  Throughput: 2298.36 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 4 workers):  1.71,  1.69,  1.74,  1.70 ms.  Throughput: 2338.12 iter/sec.
Timings for 560K FFT length (5 cores, 1 worker):  0.54 ms.  Throughput: 1835.29 iter/sec.
Timings for 560K FFT length (5 cores, 2 workers):  0.69,  0.96 ms.  Throughput: 2479.12 iter/sec.
Timings for 560K FFT length (5 cores, 3 workers):  1.91,  1.00,  1.03 ms.  Throughput: 2494.99 iter/sec.
Timings for 560K FFT length (5 cores, 4 workers):  1.92,  0.99,  1.95,  1.80 ms.  Throughput: 2598.71 iter/sec.
Timings for 560K FFT length (5 cores, 5 workers):  1.95,  1.78,  1.94,  1.96,  1.81 ms.  Throughput: 2652.21 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 1 worker):  0.45 ms.  Throughput: 2224.04 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 2 workers):  0.68,  0.93 ms.  Throughput: 2547.03 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 3 workers):  1.75,  0.99,  0.95 ms.  Throughput: 2635.47 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 4 workers):  1.75,  0.96,  1.80,  1.88 ms.  Throughput: 2703.40 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 5 workers):  1.83,  1.78,  1.79,  1.74,  1.82 ms.  Throughput: 2791.98 iter/sec.
Timings for 560K FFT length (6 cores, 1 worker):  0.46 ms.  Throughput: 2157.05 iter/sec.
Timings for 560K FFT length (6 cores, 2 workers):  0.74,  0.72 ms.  Throughput: 2738.02 iter/sec.
Timings for 560K FFT length (6 cores, 3 workers):  1.94,  1.04,  0.68 ms.  Throughput: 2954.48 iter/sec.
Timings for 560K FFT length (6 cores, 4 workers):  2.07,  1.06,  2.13,  0.96 ms.  Throughput: 2937.02 iter/sec.
Timings for 560K FFT length (6 cores, 5 workers):  1.95,  1.84,  2.12,  2.19,  0.95 ms.  Throughput: 3039.66 iter/sec.
Timings for 560K FFT length (6 cores, 6 workers):  2.03,  1.84,  2.05,  2.17,  1.83,  1.84 ms.  Throughput: 3078.81 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 1 worker):  0.39 ms.  Throughput: 2538.01 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 2 workers):  0.63,  0.61 ms.  Throughput: 3236.93 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 3 workers):  1.78,  0.89,  0.62 ms.  Throughput: 3295.60 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 4 workers):  1.83,  0.91,  1.74,  0.92 ms.  Throughput: 3304.03 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 5 workers):  1.84,  1.74,  1.81,  1.76,  0.94 ms.  Throughput: 3303.95 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 6 workers):  1.85,  1.82,  1.75,  1.82,  1.79,  1.81 ms.  Throughput: 3321.32 iter/sec.

Timings for 6144K FFT length (1 core, 1 worker): 20.81 ms.  Throughput: 48.06 iter/sec.
Timings for 6144K FFT length (1 core hyperthreaded, 1 worker): 19.92 ms.  Throughput: 50.20 iter/sec.
Timings for 6144K FFT length (2 cores, 1 worker): 10.68 ms.  Throughput: 93.67 iter/sec.
Timings for 6144K FFT length (2 cores, 2 workers): 20.90, 20.81 ms.  Throughput: 95.91 iter/sec.
Timings for 6144K FFT length (2 cores hyperthreaded, 1 worker): 10.15 ms.  Throughput: 98.49 iter/sec.
Timings for 6144K FFT length (2 cores hyperthreaded, 2 workers): 20.36, 20.25 ms.  Throughput: 98.50 iter/sec.
Timings for 6144K FFT length (3 cores, 1 worker):  7.24 ms.  Throughput: 138.13 iter/sec.
Timings for 6144K FFT length (3 cores, 2 workers): 10.76, 21.03 ms.  Throughput: 140.52 iter/sec.
Timings for 6144K FFT length (3 cores, 3 workers): 21.24, 21.09, 21.15 ms.  Throughput: 141.78 iter/sec.
Timings for 6144K FFT length (3 cores hyperthreaded, 1 worker):  6.84 ms.  Throughput: 146.29 iter/sec.
Timings for 6144K FFT length (3 cores hyperthreaded, 2 workers): 10.31, 20.37 ms.  Throughput: 146.04 iter/sec.
Timings for 6144K FFT length (3 cores hyperthreaded, 3 workers): 20.55, 20.43, 20.34 ms.  Throughput: 146.77 iter/sec.
Timings for 6144K FFT length (4 cores, 1 worker):  6.16 ms.  Throughput: 162.41 iter/sec.
Timings for 6144K FFT length (4 cores, 2 workers): 11.17, 11.16 ms.  Throughput: 179.21 iter/sec.
Timings for 6144K FFT length (4 cores, 3 workers): 22.02, 21.94, 11.31 ms.  Throughput: 179.37 iter/sec.
Timings for 6144K FFT length (4 cores, 4 workers): 22.82, 22.54, 22.99, 22.54 ms.  Throughput: 176.06 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 1 worker):  5.62 ms.  Throughput: 177.96 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 2 workers): 10.87, 10.85 ms.  Throughput: 184.11 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 3 workers): 21.73, 21.66, 11.01 ms.  Throughput: 183.04 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 4 workers): 22.09, 21.97, 22.07, 21.97 ms.  Throughput: 181.62 iter/sec.
Timings for 6144K FFT length (5 cores, 1 worker):  5.21 ms.  Throughput: 192.07 iter/sec.
Timings for 6144K FFT length (5 cores, 2 workers):  8.53, 12.85 ms.  Throughput: 195.06 iter/sec.
Timings for 6144K FFT length (5 cores, 3 workers): 25.69, 13.03, 12.51 ms.  Throughput: 195.59 iter/sec.
Timings for 6144K FFT length (5 cores, 4 workers): 26.42, 13.08, 25.31, 25.17 ms.  Throughput: 193.52 iter/sec.
Timings for 6144K FFT length (5 cores, 5 workers): 26.63, 26.25, 26.15, 25.59, 25.27 ms.  Throughput: 192.54 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 1 worker):  4.83 ms.  Throughput: 207.21 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 2 workers):  8.43, 11.93 ms.  Throughput: 202.52 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 3 workers): 25.90, 12.64, 12.08 ms.  Throughput: 200.53 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 4 workers): 26.05, 12.84, 24.32, 24.44 ms.  Throughput: 198.29 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 5 workers): 26.06, 25.97, 25.78, 24.40, 24.48 ms.  Throughput: 197.49 iter/sec.
Timings for 6144K FFT length (6 cores, 1 worker):  4.82 ms.  Throughput: 207.62 iter/sec.
Timings for 6144K FFT length (6 cores, 2 workers):  9.78,  9.60 ms.  Throughput: 206.41 iter/sec.
Timings for 6144K FFT length (6 cores, 3 workers): 29.38, 14.57,  9.74 ms.  Throughput: 205.32 iter/sec.
Timings for 6144K FFT length (6 cores, 4 workers): 29.86, 14.75, 30.39, 14.62 ms.  Throughput: 202.59 iter/sec.
Timings for 6144K FFT length (6 cores, 5 workers): 29.67, 29.64, 29.98, 30.62, 14.68 ms.  Throughput: 201.58 iter/sec.
Timings for 6144K FFT length (6 cores, 6 workers): 29.87, 29.72, 30.06, 30.51, 29.75, 29.60 ms.  Throughput: 200.57 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 1 worker):  4.60 ms.  Throughput: 217.57 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 2 workers):  9.45,  9.51 ms.  Throughput: 211.00 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 3 workers): 29.74, 14.21,  9.59 ms.  Throughput: 208.30 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 4 workers): 30.11, 14.36, 29.69, 14.74 ms.  Throughput: 204.39 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 5 workers): 30.39, 28.99, 28.93, 29.42, 14.72 ms.  Throughput: 203.91 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 6 workers): 30.59, 29.11, 28.97, 29.23, 30.76, 29.03 ms.  Throughput: 202.74 iter/sec.

And for the 3900x:

Code:

Prime95 64-bit version 30.3, RdtscTiming=1
Timings for 560K FFT length (1 core, 1 worker):  1.68 ms.  Throughput: 596.13 iter/sec.
Timings for 560K FFT length (1 core hyperthreaded, 1 worker):  1.60 ms.  Throughput: 626.46 iter/sec.
Timings for 560K FFT length (2 cores, 1 worker):  0.89 ms.  Throughput: 1124.58 iter/sec.
Timings for 560K FFT length (2 cores, 2 workers):  1.73,  1.71 ms.  Throughput: 1161.82 iter/sec.
Timings for 560K FFT length (2 cores hyperthreaded, 1 worker):  0.83 ms.  Throughput: 1199.55 iter/sec.
Timings for 560K FFT length (2 cores hyperthreaded, 2 workers):  1.64,  1.67 ms.  Throughput: 1209.46 iter/sec.
Timings for 560K FFT length (3 cores, 1 worker):  0.62 ms.  Throughput: 1622.42 iter/sec.
Timings for 560K FFT length (3 cores, 2 workers):  0.89,  1.74 ms.  Throughput: 1694.66 iter/sec.
Timings for 560K FFT length (3 cores, 3 workers):  1.71,  1.73,  1.72 ms.  Throughput: 1743.60 iter/sec.
Timings for 560K FFT length (3 cores hyperthreaded, 1 worker):  0.57 ms.  Throughput: 1749.55 iter/sec.
Timings for 560K FFT length (3 cores hyperthreaded, 2 workers):  0.85,  1.69 ms.  Throughput: 1773.91 iter/sec.
Timings for 560K FFT length (3 cores hyperthreaded, 3 workers):  1.64,  1.70,  1.64 ms.  Throughput: 1806.28 iter/sec.
Timings for 560K FFT length (4 cores, 1 worker):  0.54 ms.  Throughput: 1855.62 iter/sec.
Timings for 560K FFT length (4 cores, 2 workers):  0.91,  0.92 ms.  Throughput: 2185.70 iter/sec.
Timings for 560K FFT length (4 cores, 3 workers):  0.90,  1.71,  1.74 ms.  Throughput: 2265.57 iter/sec.
Timings for 560K FFT length (4 cores, 4 workers):  1.74,  1.72,  1.73,  1.71 ms.  Throughput: 2316.47 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 1 worker):  0.50 ms.  Throughput: 2010.02 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 2 workers):  0.86,  0.85 ms.  Throughput: 2341.66 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 3 workers):  0.86,  1.63,  1.66 ms.  Throughput: 2376.98 iter/sec.
Timings for 560K FFT length (4 cores hyperthreaded, 4 workers):  1.69,  1.64,  1.65,  1.70 ms.  Throughput: 2392.11 iter/sec.
Timings for 560K FFT length (5 cores, 1 worker):  0.52 ms.  Throughput: 1937.26 iter/sec.
Timings for 560K FFT length (5 cores, 2 workers):  0.62,  0.91 ms.  Throughput: 2713.79 iter/sec.
Timings for 560K FFT length (5 cores, 3 workers):  0.98,  0.91,  1.73 ms.  Throughput: 2698.38 iter/sec.
Timings for 560K FFT length (5 cores, 4 workers):  0.90,  1.72,  1.73,  1.74 ms.  Throughput: 2841.52 iter/sec.
Timings for 560K FFT length (5 cores, 5 workers):  1.74,  1.73,  1.71,  1.73,  1.73 ms.  Throughput: 2888.60 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 1 worker):  0.43 ms.  Throughput: 2339.93 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 2 workers):  0.60,  0.87 ms.  Throughput: 2819.06 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 3 workers):  0.85,  0.87,  1.71 ms.  Throughput: 2913.56 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 4 workers):  0.88,  1.65,  1.70,  1.71 ms.  Throughput: 2916.13 iter/sec.
Timings for 560K FFT length (5 cores hyperthreaded, 5 workers):  1.65,  1.71,  1.70,  1.71,  1.71 ms.  Throughput: 2953.25 iter/sec.
Timings for 560K FFT length (6 cores, 1 worker):  0.42 ms.  Throughput: 2387.22 iter/sec.
Timings for 560K FFT length (6 cores, 2 workers):  0.66,  0.62 ms.  Throughput: 3131.41 iter/sec.
Timings for 560K FFT length (6 cores, 3 workers):  0.90,  0.97,  0.91 ms.  Throughput: 3230.09 iter/sec.
Timings for 560K FFT length (6 cores, 4 workers):  0.90,  0.92,  1.73,  1.74 ms.  Throughput: 3347.15 iter/sec.
Timings for 560K FFT length (6 cores, 5 workers):  1.73,  1.72,  0.90,  1.74,  1.73 ms.  Throughput: 3423.57 iter/sec.
Timings for 560K FFT length (6 cores, 6 workers):  1.73,  1.73,  1.73,  1.72,  1.73,  1.73 ms.  Throughput: 3470.47 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 1 worker):  0.36 ms.  Throughput: 2800.22 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 2 workers):  0.58,  0.58 ms.  Throughput: 3429.08 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 3 workers):  0.88,  0.86,  0.88 ms.  Throughput: 3433.78 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 4 workers):  0.86,  0.86,  1.72,  1.71 ms.  Throughput: 3497.29 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 5 workers):  1.72,  1.65,  0.86,  1.67,  1.65 ms.  Throughput: 3556.63 iter/sec.
Timings for 560K FFT length (6 cores hyperthreaded, 6 workers):  1.71,  1.71,  1.66,  1.71,  1.65,  1.71 ms.  Throughput: 3548.46 iter/sec.
Timings for 560K FFT length (7 cores, 1 worker):  0.38 ms.  Throughput: 2603.88 iter/sec.
Timings for 560K FFT length (7 cores, 2 workers):  0.59,  0.68 ms.  Throughput: 3150.85 iter/sec.
Timings for 560K FFT length (7 cores, 3 workers):  0.63,  0.99,  0.99 ms.  Throughput: 3603.79 iter/sec.
Timings for 560K FFT length (7 cores, 4 workers):  0.92,  0.92,  0.96,  1.74 ms.  Throughput: 3786.51 iter/sec.
Timings for 560K FFT length (7 cores, 5 workers):  1.74,  1.74,  0.91,  0.97,  1.74 ms.  Throughput: 3849.78 iter/sec.
Timings for 560K FFT length (7 cores, 6 workers):  1.73,  1.74,  1.75,  1.74,  0.91,  1.74 ms.  Throughput: 3970.95 iter/sec.
Timings for 560K FFT length (7 cores, 7 workers):  1.75,  1.73,  1.75,  1.74,  1.75,  1.75,  1.76 ms.  Throughput: 4011.95 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 1 worker):  0.35 ms.  Throughput: 2843.81 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 2 workers):  0.51,  0.60 ms.  Throughput: 3657.01 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 3 workers):  0.59,  0.88,  0.86 ms.  Throughput: 4006.25 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 4 workers):  0.90,  0.88,  0.90,  1.66 ms.  Throughput: 3969.22 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 5 workers):  1.74,  1.69,  0.87,  0.86,  1.67 ms.  Throughput: 4078.86 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 6 workers):  1.67,  1.73,  1.68,  1.66,  0.88,  1.68 ms.  Throughput: 4104.73 iter/sec.
Timings for 560K FFT length (7 cores hyperthreaded, 7 workers):  1.73,  1.67,  1.73,  1.73,  1.67,  1.66,  1.73 ms.  Throughput: 4115.35 iter/sec.
Timings for 560K FFT length (8 cores, 1 worker):  0.38 ms.  Throughput: 2656.69 iter/sec.
Timings for 560K FFT length (8 cores, 2 workers):  0.61,  0.60 ms.  Throughput: 3307.63 iter/sec.
Timings for 560K FFT length (8 cores, 3 workers):  0.67,  0.67,  0.92 ms.  Throughput: 4056.22 iter/sec.
Timings for 560K FFT length (8 cores, 4 workers):  0.93,  0.98,  0.94,  0.91 ms.  Throughput: 4258.06 iter/sec.
Timings for 560K FFT length (8 cores, 5 workers):  1.77,  1.74,  0.93,  0.92,  0.92 ms.  Throughput: 4392.24 iter/sec.
Timings for 560K FFT length (8 cores, 6 workers):  1.76,  1.75,  1.75,  1.74,  0.92,  0.92 ms.  Throughput: 4466.09 iter/sec.
Timings for 560K FFT length (8 cores, 7 workers):  1.76,  1.74,  1.77,  1.74,  1.76,  1.78,  0.91 ms.  Throughput: 4505.90 iter/sec.
Timings for 560K FFT length (8 cores, 8 workers):  1.76,  1.76,  1.73,  1.76,  1.76,  1.76,  1.75,  1.74 ms.  Throughput: 4566.36 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 1 worker):  0.35 ms.  Throughput: 2881.36 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 2 workers):  0.52,  0.49 ms.  Throughput: 3998.15 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 3 workers):  0.60,  0.62,  0.91 ms.  Throughput: 4377.85 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 4 workers):  0.92,  0.90,  0.91,  0.92 ms.  Throughput: 4377.48 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 5 workers):  1.75,  1.69,  0.90,  0.88,  0.89 ms.  Throughput: 4533.52 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 6 workers):  1.72,  1.72,  1.68,  1.71,  0.88,  0.90 ms.  Throughput: 4598.71 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 7 workers):  1.70,  1.69,  1.70,  1.69,  1.78,  1.75,  0.89 ms.  Throughput: 4617.43 iter/sec.
Timings for 560K FFT length (8 cores hyperthreaded, 8 workers):  1.70,  1.71,  1.69,  1.75,  1.75,  1.75,  1.75,  1.69 ms.  Throughput: 4645.92 iter/sec.
Timings for 560K FFT length (9 cores, 1 worker):  0.35 ms.  Throughput: 2840.46 iter/sec.
Timings for 560K FFT length (9 cores, 2 workers):  0.51,  0.60 ms.  Throughput: 3629.79 iter/sec.
Timings for 560K FFT length (9 cores, 3 workers):  0.70,  0.70,  0.63 ms.  Throughput: 4437.90 iter/sec.
Timings for 560K FFT length (9 cores, 4 workers):  0.67,  0.99,  0.92,  0.94 ms.  Throughput: 4653.58 iter/sec.
Timings for 560K FFT length (9 cores, 5 workers):  1.76,  0.99,  0.93,  0.99,  0.99 ms.  Throughput: 4671.26 iter/sec.
Timings for 560K FFT length (9 cores, 6 workers):  1.76,  0.96,  1.75,  1.78,  0.96,  0.95 ms.  Throughput: 4835.69 iter/sec.
Timings for 560K FFT length (9 cores, 7 workers):  1.76,  0.99,  1.75,  1.75,  1.75,  1.78,  0.93 ms.  Throughput: 4934.32 iter/sec.
Timings for 560K FFT length (9 cores, 8 workers):  1.78,  0.92,  1.78,  1.76,  1.76,  1.75,  1.76,  1.76 ms.  Throughput: 5047.05 iter/sec.
Timings for 560K FFT length (9 cores, 9 workers):  1.79,  1.76,  1.77,  1.76,  1.76,  1.75,  1.77,  1.77,  1.76 ms.  Throughput: 5095.59 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 1 worker):  0.46 ms.  Throughput: 2193.08 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 2 workers):  0.46,  0.51 ms.  Throughput: 4120.53 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 3 workers):  0.64,  0.64,  0.63 ms.  Throughput: 4702.55 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 4 workers):  0.61,  0.92,  0.92,  0.90 ms.  Throughput: 4918.19 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 5 workers):  1.73,  0.92,  0.93,  0.93,  0.89 ms.  Throughput: 4947.98 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 6 workers):  1.80,  0.92,  1.79,  1.72,  0.90,  0.90 ms.  Throughput: 4992.64 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 7 workers):  1.79,  0.90,  1.73,  1.80,  1.73,  1.80,  0.90 ms.  Throughput: 5045.36 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 8 workers):  1.75,  0.90,  1.80,  1.73,  1.80,  1.74,  1.75,  1.80 ms.  Throughput: 5072.50 iter/sec.
Timings for 560K FFT length (9 cores hyperthreaded, 9 workers):  1.81,  1.76,  1.82,  1.73,  1.75,  1.82,  1.75,  1.72,  1.82 ms.  Throughput: 5073.20 iter/sec.
Timings for 560K FFT length (10 cores, 1 worker):  0.35 ms.  Throughput: 2828.22 iter/sec.
Timings for 560K FFT length (10 cores, 2 workers):  0.50,  0.51 ms.  Throughput: 3966.05 iter/sec.
Timings for 560K FFT length (10 cores, 3 workers):  0.60,  0.69,  0.64 ms.  Throughput: 4678.47 iter/sec.
Timings for 560K FFT length (10 cores, 4 workers):  0.67,  0.67,  0.99,  0.99 ms.  Throughput: 4992.45 iter/sec.
Timings for 560K FFT length (10 cores, 5 workers):  1.80,  0.94,  0.67,  1.01,  0.93 ms.  Throughput: 5179.04 iter/sec.
Timings for 560K FFT length (10 cores, 6 workers):  1.77,  0.97,  1.77,  0.94,  0.95,  0.93 ms.  Throughput: 5340.94 iter/sec.
Timings for 560K FFT length (10 cores, 7 workers):  1.79,  1.00,  1.79,  0.93,  1.79,  1.77,  1.00 ms.  Throughput: 5313.61 iter/sec.
Timings for 560K FFT length (10 cores, 8 workers):  1.80,  0.96,  1.79,  0.94,  1.80,  1.78,  1.78,  1.77 ms.  Throughput: 5471.75 iter/sec.
Timings for 560K FFT length (10 cores, 9 workers):  1.79,  1.82,  1.77,  1.79,  0.94,  1.79,  1.78,  1.79,  1.79 ms.  Throughput: 5533.93 iter/sec.
Timings for 560K FFT length (10 cores, 10 workers):  1.80,  1.79,  1.79,  1.79,  1.77,  1.79,  1.78,  1.79,  1.79,  1.78 ms.  Throughput: 5597.60 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 1 worker):  0.53 ms.  Throughput: 1883.69 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 2 workers):  0.47,  0.45 ms.  Throughput: 4337.74 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 3 workers):  0.52,  0.62,  0.63 ms.  Throughput: 5113.00 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 4 workers):  0.64,  0.62,  0.91,  0.91 ms.  Throughput: 5370.38 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 5 workers):  1.82,  0.93,  0.62,  0.93,  0.94 ms.  Throughput: 5366.16 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 6 workers):  1.78,  0.93,  1.82,  0.91,  0.91,  0.92 ms.  Throughput: 5483.01 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 7 workers):  1.77,  0.96,  1.75,  0.94,  1.83,  1.76,  0.92 ms.  Throughput: 5439.21 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 8 workers):  1.77,  0.92,  1.82,  0.92,  1.83,  1.84,  1.75,  1.83 ms.  Throughput: 5489.56 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 9 workers):  1.82,  1.82,  1.77,  1.79,  0.91,  1.82,  1.77,  1.82,  1.77 ms.  Throughput: 5552.93 iter/sec.
Timings for 560K FFT length (10 cores hyperthreaded, 10 workers):  1.77,  1.76,  1.85,  1.78,  1.85,  1.85,  1.85,  1.79,  1.76,  1.80 ms.  Throughput: 5540.64 iter/sec.
Timings for 560K FFT length (11 cores, 1 worker):  0.36 ms.  Throughput: 2799.04 iter/sec.
Timings for 560K FFT length (11 cores, 2 workers):  0.48,  0.51 ms.  Throughput: 4047.81 iter/sec.
Timings for 560K FFT length (11 cores, 3 workers):  0.58,  0.63,  0.75 ms.  Throughput: 4657.55 iter/sec.
Timings for 560K FFT length (11 cores, 4 workers):  0.69,  0.69,  0.66,  0.96 ms.  Throughput: 5458.38 iter/sec.
Timings for 560K FFT length (11 cores, 5 workers):  1.78,  1.03,  0.69,  0.69,  1.03 ms.  Throughput: 5394.65 iter/sec.
Timings for 560K FFT length (11 cores, 6 workers):  1.80,  1.03,  1.95,  0.98,  0.70,  1.03 ms.  Throughput: 5454.38 iter/sec.
Timings for 560K FFT length (11 cores, 7 workers):  1.81,  0.96,  1.85,  1.02,  1.81,  0.97,  1.01 ms.  Throughput: 5699.98 iter/sec.
Timings for 560K FFT length (11 cores, 8 workers):  1.83,  0.95,  1.87,  0.95,  1.82,  0.97,  1.83,  1.81 ms.  Throughput: 5859.08 iter/sec.
Timings for 560K FFT length (11 cores, 9 workers):  1.81,  1.83,  1.81,  1.84,  0.96,  1.80,  0.95,  1.82,  1.83 ms.  Throughput: 5938.98 iter/sec.
Timings for 560K FFT length (11 cores, 10 workers):  1.82,  1.81,  1.83,  1.82,  1.87,  1.82,  1.83,  0.96,  1.82,  1.82 ms.  Throughput: 5974.92 iter/sec.
Timings for 560K FFT length (11 cores, 11 workers):  1.82,  1.83,  1.81,  1.86,  1.88,  1.83,  1.83,  1.84,  1.81,  1.82,  1.82 ms.  Throughput: 6004.05 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 1 worker):  0.58 ms.  Throughput: 1736.28 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 2 workers):  0.42,  0.48 ms.  Throughput: 4461.85 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 3 workers):  0.54,  0.64,  0.67 ms.  Throughput: 4925.70 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 4 workers):  0.65,  0.66,  0.65,  0.97 ms.  Throughput: 5625.17 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 5 workers):  1.80,  0.95,  0.65,  0.66,  0.97 ms.  Throughput: 5715.23 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 6 workers):  1.85,  0.94,  1.79,  0.95,  0.65,  0.96 ms.  Throughput: 5790.92 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 7 workers):  1.86,  0.92,  1.84,  0.96,  1.89,  0.95,  0.95 ms.  Throughput: 5843.95 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 8 workers):  1.81,  0.95,  1.86,  0.94,  1.80,  0.95,  1.86,  1.87 ms.  Throughput: 5887.58 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 9 workers):  1.86,  1.81,  1.85,  1.87,  0.96,  1.87,  0.94,  1.79,  1.83 ms.  Throughput: 5904.31 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 10 workers):  1.86,  1.80,  1.80,  1.80,  1.80,  1.86,  1.84,  0.97,  1.81,  1.81 ms.  Throughput: 5975.83 iter/sec.
Timings for 560K FFT length (11 cores hyperthreaded, 11 workers):  1.81,  1.86,  1.81,  1.82,  1.81,  1.81,  1.82,  1.82,  1.89,  1.80,  1.82 ms.  Throughput: 6026.30 iter/sec.
Timings for 560K FFT length (12 cores, 1 worker):  0.35 ms.  Throughput: 2836.34 iter/sec.
Timings for 560K FFT length (12 cores, 2 workers):  0.49,  0.47 ms.  Throughput: 4180.80 iter/sec.
Timings for 560K FFT length (12 cores, 3 workers):  0.70,  0.63,  0.68 ms.  Throughput: 4494.87 iter/sec.
Timings for 560K FFT length (12 cores, 4 workers):  0.69,  0.75,  0.72,  0.73 ms.  Throughput: 5536.42 iter/sec.
Timings for 560K FFT length (12 cores, 5 workers):  1.82,  1.03,  0.68,  0.67,  0.72 ms.  Throughput: 5866.73 iter/sec.
Timings for 560K FFT length (12 cores, 6 workers):  1.85,  0.98,  1.87,  1.05,  0.71,  0.71 ms.  Throughput: 5863.54 iter/sec.
Timings for 560K FFT length (12 cores, 7 workers):  1.86,  0.98,  1.89,  1.04,  1.86,  0.98,  0.67 ms.  Throughput: 6087.19 iter/sec.
Timings for 560K FFT length (12 cores, 8 workers):  1.84,  1.03,  1.91,  0.96,  1.85,  1.03,  1.85,  1.05 ms.  Throughput: 6074.77 iter/sec.
Timings for 560K FFT length (12 cores, 9 workers):  1.84,  1.87,  1.85,  1.91,  1.05,  1.83,  0.97,  1.84,  1.07 ms.  Throughput: 6149.46 iter/sec.
Timings for 560K FFT length (12 cores, 10 workers):  1.86,  1.85,  1.86,  1.86,  1.91,  1.84,  1.85,  0.97,  1.84,  1.05 ms.  Throughput: 6284.85 iter/sec.
Timings for 560K FFT length (12 cores, 11 workers):  1.91,  1.88,  1.86,  1.90,  1.96,  1.86,  1.86,  1.87,  1.86,  1.86,  1.03 ms.  Throughput: 6279.52 iter/sec.
Timings for 560K FFT length (12 cores, 12 workers):  1.99,  1.88,  1.87,  1.91,  1.91,  1.85,  1.87,  1.88,  1.86,  1.87,  1.92,  2.39 ms.  Throughput: 6235.11 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 1 worker):  0.47 ms.  Throughput: 2146.80 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 2 workers):  0.44,  0.45 ms.  Throughput: 4467.42 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 3 workers):  0.67,  0.60,  0.72 ms.  Throughput: 4540.16 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 4 workers):  0.66,  0.66,  0.65,  0.68 ms.  Throughput: 6048.21 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 5 workers):  1.91,  0.98,  0.67,  0.67,  0.68 ms.  Throughput: 6009.17 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 6 workers):  1.93,  0.94,  1.85,  0.97,  0.66,  0.68 ms.  Throughput: 6130.32 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 7 workers):  1.89,  0.99,  1.90,  1.02,  1.91,  0.98,  0.67 ms.  Throughput: 6092.31 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 8 workers):  1.84,  0.97,  1.92,  0.98,  1.85,  1.00,  1.92,  1.01 ms.  Throughput: 6174.14 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 9 workers):  1.92,  1.85,  1.88,  1.87,  1.02,  1.90,  0.97,  1.90,  0.96 ms.  Throughput: 6231.03 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 10 workers):  1.93,  1.91,  1.92,  1.90,  1.90,  1.85,  1.90,  1.00,  1.84,  1.02 ms.  Throughput: 6196.80 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 11 workers):  1.94,  1.92,  1.86,  1.91,  1.92,  1.92,  1.92,  1.92,  1.91,  1.89,  1.00 ms.  Throughput: 6230.77 iter/sec.
Timings for 560K FFT length (12 cores hyperthreaded, 12 workers):  1.95,  1.84,  1.94,  1.90,  1.90,  1.93,  1.92,  1.94,  1.94,  1.85,  1.96,  1.92 ms.  Throughput: 6265.96 iter/sec.

Timings for 6144K FFT length (1 core, 1 worker): 20.25 ms.  Throughput: 49.39 iter/sec.
Timings for 6144K FFT length (1 core hyperthreaded, 1 worker): 19.34 ms.  Throughput: 51.71 iter/sec.
Timings for 6144K FFT length (2 cores, 1 worker): 10.49 ms.  Throughput: 95.31 iter/sec.
Timings for 6144K FFT length (2 cores, 2 workers): 20.73, 20.63 ms.  Throughput: 96.72 iter/sec.
Timings for 6144K FFT length (2 cores hyperthreaded, 1 worker):  9.92 ms.  Throughput: 100.78 iter/sec.
Timings for 6144K FFT length (2 cores hyperthreaded, 2 workers): 19.95, 19.94 ms.  Throughput: 100.27 iter/sec.
Timings for 6144K FFT length (3 cores, 1 worker):  7.07 ms.  Throughput: 141.42 iter/sec.
Timings for 6144K FFT length (3 cores, 2 workers): 10.62, 20.74 ms.  Throughput: 142.35 iter/sec.
Timings for 6144K FFT length (3 cores, 3 workers): 20.75, 20.78, 20.74 ms.  Throughput: 144.52 iter/sec.
Timings for 6144K FFT length (3 cores hyperthreaded, 1 worker):  6.74 ms.  Throughput: 148.37 iter/sec.
Timings for 6144K FFT length (3 cores hyperthreaded, 2 workers): 10.15, 20.03 ms.  Throughput: 148.45 iter/sec.
Timings for 6144K FFT length (3 cores hyperthreaded, 3 workers): 20.08, 19.88, 20.30 ms.  Throughput: 149.35 iter/sec.
Timings for 6144K FFT length (4 cores, 1 worker):  6.02 ms.  Throughput: 166.15 iter/sec.
Timings for 6144K FFT length (4 cores, 2 workers): 11.06, 11.03 ms.  Throughput: 181.03 iter/sec.
Timings for 6144K FFT length (4 cores, 3 workers): 11.11, 21.69, 21.69 ms.  Throughput: 182.25 iter/sec.
Timings for 6144K FFT length (4 cores, 4 workers): 21.72, 21.79, 21.79, 21.72 ms.  Throughput: 183.86 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 1 worker):  5.36 ms.  Throughput: 186.59 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 2 workers): 10.76, 10.76 ms.  Throughput: 185.88 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 3 workers): 10.68, 21.58, 21.58 ms.  Throughput: 186.35 iter/sec.
Timings for 6144K FFT length (4 cores hyperthreaded, 4 workers): 21.60, 21.34, 21.04, 21.61 ms.  Throughput: 186.96 iter/sec.
Timings for 6144K FFT length (5 cores, 1 worker):  5.11 ms.  Throughput: 195.52 iter/sec.
Timings for 6144K FFT length (5 cores, 2 workers):  8.28, 12.78 ms.  Throughput: 199.03 iter/sec.
Timings for 6144K FFT length (5 cores, 3 workers): 12.10, 12.53, 25.07 ms.  Throughput: 202.37 iter/sec.
Timings for 6144K FFT length (5 cores, 4 workers): 12.67, 24.36, 24.12, 24.05 ms.  Throughput: 203.05 iter/sec.
Timings for 6144K FFT length (5 cores, 5 workers): 25.47, 25.18, 24.70, 24.22, 24.22 ms.  Throughput: 202.05 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 1 worker):  4.70 ms.  Throughput: 212.57 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 2 workers):  8.16, 11.83 ms.  Throughput: 207.00 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 3 workers): 11.89, 12.17, 24.57 ms.  Throughput: 206.98 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 4 workers): 12.13, 24.39, 23.31, 24.13 ms.  Throughput: 207.78 iter/sec.
Timings for 6144K FFT length (5 cores hyperthreaded, 5 workers): 24.82, 24.82, 24.56, 24.40, 23.66 ms.  Throughput: 204.55 iter/sec.
Timings for 6144K FFT length (6 cores, 1 worker):  4.70 ms.  Throughput: 212.77 iter/sec.
Timings for 6144K FFT length (6 cores, 2 workers):  9.44,  9.41 ms.  Throughput: 212.25 iter/sec.
Timings for 6144K FFT length (6 cores, 3 workers): 13.66, 14.69, 14.20 ms.  Throughput: 211.71 iter/sec.
Timings for 6144K FFT length (6 cores, 4 workers): 14.35, 14.12, 27.29, 27.34 ms.  Throughput: 213.70 iter/sec.
Timings for 6144K FFT length (6 cores, 5 workers): 29.14, 28.88, 14.54, 27.29, 27.29 ms.  Throughput: 211.03 iter/sec.
Timings for 6144K FFT length (6 cores, 6 workers): 29.31, 29.00, 29.11, 28.93, 27.90, 27.87 ms.  Throughput: 209.24 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 1 worker):  4.46 ms.  Throughput: 224.38 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 2 workers):  9.19,  9.19 ms.  Throughput: 217.52 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 3 workers): 12.96, 14.26, 14.08 ms.  Throughput: 218.29 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 4 workers): 14.23, 14.10, 26.07, 26.07 ms.  Throughput: 217.91 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 5 workers): 28.70, 28.72, 14.33, 26.03, 26.03 ms.  Throughput: 216.25 iter/sec.
Timings for 6144K FFT length (6 cores hyperthreaded, 6 workers): 29.26, 29.51, 29.87, 28.92, 25.92, 25.76 ms.  Throughput: 213.52 iter/sec.
Timings for 6144K FFT length (7 cores, 1 worker):  4.01 ms.  Throughput: 249.23 iter/sec.
Timings for 6144K FFT length (7 cores, 2 workers):  7.69,  9.90 ms.  Throughput: 231.01 iter/sec.
Timings for 6144K FFT length (7 cores, 3 workers): 10.30, 16.60, 16.60 ms.  Throughput: 217.54 iter/sec.
Timings for 6144K FFT length (7 cores, 4 workers): 16.05, 16.40, 15.73, 30.40 ms.  Throughput: 219.75 iter/sec.
Timings for 6144K FFT length (7 cores, 5 workers): 32.84, 32.82, 16.13, 15.85, 30.67 ms.  Throughput: 218.59 iter/sec.
Timings for 6144K FFT length (7 cores, 6 workers): 33.33, 33.33, 33.33, 33.33, 15.78, 30.97 ms.  Throughput: 215.66 iter/sec.
Timings for 6144K FFT length (7 cores, 7 workers): 33.66, 33.64, 33.66, 33.66, 31.85, 31.85, 31.32 ms.  Throughput: 213.58 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 1 worker):  3.81 ms.  Throughput: 262.58 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 2 workers):  7.22,  9.76 ms.  Throughput: 241.02 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 3 workers): 10.07, 16.31, 16.31 ms.  Throughput: 221.94 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 4 workers): 16.54, 16.37, 15.34, 28.45 ms.  Throughput: 221.89 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 5 workers): 34.02, 33.31, 16.61, 15.46, 29.11 ms.  Throughput: 218.66 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 6 workers): 33.97, 33.98, 34.13, 34.13, 15.44, 29.17 ms.  Throughput: 216.52 iter/sec.
Timings for 6144K FFT length (7 cores hyperthreaded, 7 workers): 34.53, 34.53, 34.62, 34.62, 30.83, 30.94, 29.48 ms.  Throughput: 214.37 iter/sec.
Timings for 6144K FFT length (8 cores, 1 worker):  3.70 ms.  Throughput: 270.56 iter/sec.
Timings for 6144K FFT length (8 cores, 2 workers):  8.05,  8.05 ms.  Throughput: 248.42 iter/sec.
Timings for 6144K FFT length (8 cores, 3 workers): 11.28, 12.87, 17.65 ms.  Throughput: 223.06 iter/sec.
Timings for 6144K FFT length (8 cores, 4 workers): 18.06, 18.06, 18.07, 18.06 ms.  Throughput: 221.42 iter/sec.
Timings for 6144K FFT length (8 cores, 5 workers): 36.43, 36.06, 18.07, 18.21, 18.10 ms.  Throughput: 220.68 iter/sec.
Timings for 6144K FFT length (8 cores, 6 workers): 36.88, 36.88, 36.73, 36.64, 17.97, 18.20 ms.  Throughput: 219.35 iter/sec.
Timings for 6144K FFT length (8 cores, 7 workers): 37.21, 37.21, 37.10, 37.19, 36.46, 36.47, 18.26 ms.  Throughput: 217.21 iter/sec.
Timings for 6144K FFT length (8 cores, 8 workers): 37.56, 37.56, 37.59, 37.58, 37.03, 37.04, 37.01, 37.01 ms.  Throughput: 214.50 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 1 worker):  3.56 ms.  Throughput: 280.78 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 2 workers):  7.73,  7.85 ms.  Throughput: 256.68 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 3 workers): 10.49, 14.07, 17.96 ms.  Throughput: 222.04 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 4 workers): 17.97, 18.14, 17.98, 18.10 ms.  Throughput: 221.62 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 5 workers): 37.50, 36.39, 18.18, 18.19, 18.20 ms.  Throughput: 219.05 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 6 workers): 37.40, 37.36, 36.75, 37.62, 18.17, 18.30 ms.  Throughput: 216.98 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 7 workers): 37.70, 37.57, 37.77, 37.68, 37.27, 36.52, 18.29 ms.  Throughput: 215.04 iter/sec.
Timings for 6144K FFT length (8 cores hyperthreaded, 8 workers): 37.59, 37.81, 37.76, 37.76, 37.52, 37.52, 37.55, 37.55 ms.  Throughput: 212.58 iter/sec.
Timings for 6144K FFT length (9 cores, 1 worker):  3.24 ms.  Throughput: 308.62 iter/sec.
Timings for 6144K FFT length (9 cores, 2 workers):  7.43,  8.58 ms.  Throughput: 251.14 iter/sec.
Timings for 6144K FFT length (9 cores, 3 workers): 11.60, 14.53, 14.31 ms.  Throughput: 224.92 iter/sec.
Timings for 6144K FFT length (9 cores, 4 workers): 14.56, 19.98, 19.27, 18.74 ms.  Throughput: 223.99 iter/sec.
Timings for 6144K FFT length (9 cores, 5 workers): 44.13, 22.26, 20.17, 19.01, 19.42 ms.  Throughput: 221.26 iter/sec.
Timings for 6144K FFT length (9 cores, 6 workers): 44.83, 22.37, 40.98, 40.98, 19.44, 19.07 ms.  Throughput: 219.69 iter/sec.
Timings for 6144K FFT length (9 cores, 7 workers): 45.38, 22.46, 41.47, 41.46, 38.48, 38.58, 19.42 ms.  Throughput: 218.19 iter/sec.
Timings for 6144K FFT length (9 cores, 8 workers): 45.36, 22.85, 41.68, 41.68, 39.24, 39.25, 39.21, 39.21 ms.  Throughput: 215.77 iter/sec.
Timings for 6144K FFT length (9 cores, 9 workers): 46.10, 45.65, 45.65, 41.35, 41.46, 39.17, 39.28, 39.43, 39.43 ms.  Throughput: 215.52 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 1 worker):  3.14 ms.  Throughput: 318.25 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 2 workers):  7.72,  8.10 ms.  Throughput: 252.89 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 3 workers): 10.71, 15.61, 15.55 ms.  Throughput: 221.69 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 4 workers): 15.98, 19.81, 18.70, 18.69 ms.  Throughput: 220.03 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 5 workers): 50.69, 24.51, 19.73, 18.52, 18.74 ms.  Throughput: 218.58 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 6 workers): 51.27, 24.67, 40.66, 40.67, 18.55, 18.74 ms.  Throughput: 216.48 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 7 workers): 50.78, 25.39, 40.94, 40.28, 38.40, 37.59, 18.88 ms.  Throughput: 213.95 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 8 workers): 51.02, 25.37, 41.88, 40.47, 38.40, 38.38, 38.36, 38.44 ms.  Throughput: 211.78 iter/sec.
Timings for 6144K FFT length (9 cores hyperthreaded, 9 workers): 50.98, 51.00, 51.01, 40.91, 40.92, 39.48, 38.35, 38.07, 38.31 ms.  Throughput: 211.48 iter/sec.
Timings for 6144K FFT length (10 cores, 1 worker):  2.96 ms.  Throughput: 337.69 iter/sec.
Timings for 6144K FFT length (10 cores, 2 workers):  7.90,  7.90 ms.  Throughput: 253.28 iter/sec.
Timings for 6144K FFT length (10 cores, 3 workers):  9.22, 15.45, 15.63 ms.  Throughput: 237.12 iter/sec.
Timings for 6144K FFT length (10 cores, 4 workers): 16.18, 16.22, 19.99, 20.00 ms.  Throughput: 223.48 iter/sec.
Timings for 6144K FFT length (10 cores, 5 workers): 49.20, 25.27, 16.59, 19.78, 19.84 ms.  Throughput: 221.11 iter/sec.
Timings for 6144K FFT length (10 cores, 6 workers): 50.84, 25.36, 50.17, 25.57, 19.72, 19.81 ms.  Throughput: 219.34 iter/sec.
Timings for 6144K FFT length (10 cores, 7 workers): 51.44, 25.57, 50.86, 25.82, 39.63, 39.63, 19.82 ms.  Throughput: 217.87 iter/sec.
Timings for 6144K FFT length (10 cores, 8 workers): 50.37, 25.57, 51.14, 25.16, 40.56, 40.89, 41.16, 41.19 ms.  Throughput: 215.94 iter/sec.
Timings for 6144K FFT length (10 cores, 9 workers): 52.10, 51.54, 51.54, 50.96, 25.87, 40.14, 40.26, 40.43, 40.43 ms.  Throughput: 215.49 iter/sec.
Timings for 6144K FFT length (10 cores, 10 workers): 52.66, 51.87, 51.88, 51.76, 51.94, 51.76, 40.21, 40.21, 40.29, 40.29 ms.  Throughput: 214.82 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 1 worker):  2.80 ms.  Throughput: 356.54 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 2 workers):  8.07,  8.08 ms.  Throughput: 247.65 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 3 workers):  8.29, 17.10, 17.10 ms.  Throughput: 237.66 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 4 workers): 17.52, 17.54, 18.84, 18.93 ms.  Throughput: 219.99 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 5 workers): 54.03, 27.02, 17.63, 19.02, 18.75 ms.  Throughput: 218.16 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 6 workers): 54.60, 27.29, 54.59, 27.28, 18.97, 18.99 ms.  Throughput: 215.33 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 7 workers): 54.81, 27.35, 54.76, 27.37, 38.71, 38.71, 19.01 ms.  Throughput: 213.89 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 8 workers): 54.59, 27.79, 56.11, 27.22, 38.97, 38.86, 38.84, 38.74 ms.  Throughput: 211.82 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 9 workers): 55.00, 57.50, 55.00, 56.10, 27.48, 38.75, 38.88, 38.73, 38.83 ms.  Throughput: 211.07 iter/sec.
Timings for 6144K FFT length (10 cores hyperthreaded, 10 workers): 55.70, 55.71, 55.72, 56.82, 54.98, 54.97, 38.83, 39.17, 38.77, 38.88 ms.  Throughput: 210.63 iter/sec.
Timings for 6144K FFT length (11 cores, 1 worker):  2.79 ms.  Throughput: 358.36 iter/sec.
Timings for 6144K FFT length (11 cores, 2 workers):  7.60,  8.34 ms.  Throughput: 251.53 iter/sec.
Timings for 6144K FFT length (11 cores, 3 workers):  9.24, 13.72, 18.36 ms.  Throughput: 235.56 iter/sec.
Timings for 6144K FFT length (11 cores, 4 workers): 17.23, 17.49, 16.09, 21.55 ms.  Throughput: 223.74 iter/sec.
Timings for 6144K FFT length (11 cores, 5 workers): 52.67, 26.61, 17.70, 16.15, 21.70 ms.  Throughput: 221.06 iter/sec.
Timings for 6144K FFT length (11 cores, 6 workers): 53.70, 26.97, 54.21, 26.86, 16.16, 21.63 ms.  Throughput: 219.48 iter/sec.
Timings for 6144K FFT length (11 cores, 7 workers): 54.14, 27.22, 53.71, 27.45, 48.86, 24.46, 22.13 ms.  Throughput: 216.79 iter/sec.
Timings for 6144K FFT length (11 cores, 8 workers): 54.60, 27.15, 54.15, 27.57, 48.43, 25.02, 44.82, 44.83 ms.  Throughput: 215.12 iter/sec.
Timings for 6144K FFT length (11 cores, 9 workers): 55.11, 54.56, 54.51, 54.97, 27.32, 48.25, 25.04, 44.60, 44.61 ms.  Throughput: 215.11 iter/sec.
Timings for 6144K FFT length (11 cores, 10 workers): 55.51, 55.10, 55.08, 54.72, 55.36, 54.52, 48.16, 25.07, 44.63, 44.63 ms.  Throughput: 214.46 iter/sec.
Timings for 6144K FFT length (11 cores, 11 workers): 55.62, 55.20, 55.19, 55.64, 55.51, 55.11, 49.55, 49.95, 49.78, 43.96, 44.11 ms.  Throughput: 214.06 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 1 worker):  2.56 ms.  Throughput: 390.79 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 2 workers):  8.00,  8.23 ms.  Throughput: 246.62 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 3 workers):  8.30, 15.00, 20.98 ms.  Throughput: 234.76 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 4 workers): 18.08, 18.14, 16.41, 20.83 ms.  Throughput: 219.40 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 5 workers): 57.18, 27.62, 18.17, 16.31, 21.08 ms.  Throughput: 217.47 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 6 workers): 57.97, 27.69, 56.72, 28.03, 16.27, 20.89 ms.  Throughput: 216.01 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 7 workers): 56.58, 28.21, 56.46, 28.22, 51.50, 25.09, 21.06 ms.  Throughput: 213.03 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 8 workers): 56.80, 28.16, 57.47, 28.10, 51.79, 25.72, 42.51, 42.73 ms.  Throughput: 211.21 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 9 workers): 59.35, 56.06, 56.06, 56.50, 28.30, 51.82, 25.90, 42.17, 42.94 ms.  Throughput: 210.48 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 10 workers): 56.55, 58.39, 56.33, 57.01, 57.02, 56.78, 51.68, 25.83, 42.36, 42.49 ms.  Throughput: 210.44 iter/sec.
Timings for 6144K FFT length (11 cores hyperthreaded, 11 workers): 57.01, 56.79, 57.22, 56.34, 56.34, 58.50, 51.65, 52.25, 52.37, 42.54, 42.67 ms.  Throughput: 209.76 iter/sec.
Timings for 6144K FFT length (12 cores, 1 worker):  2.66 ms.  Throughput: 376.21 iter/sec.
Timings for 6144K FFT length (12 cores, 2 workers):  8.02,  8.02 ms.  Throughput: 249.25 iter/sec.
Timings for 6144K FFT length (12 cores, 3 workers): 10.63, 14.22, 15.08 ms.  Throughput: 230.72 iter/sec.
Timings for 6144K FFT length (12 cores, 4 workers): 18.00, 18.06, 17.90, 17.96 ms.  Throughput: 222.47 iter/sec.
Timings for 6144K FFT length (12 cores, 5 workers): 54.07, 27.65, 18.31, 17.92, 18.15 ms.  Throughput: 220.16 iter/sec.
Timings for 6144K FFT length (12 cores, 6 workers): 55.97, 27.72, 55.37, 28.26, 18.00, 18.01 ms.  Throughput: 218.46 iter/sec.
Timings for 6144K FFT length (12 cores, 7 workers): 55.51, 28.22, 55.47, 28.28, 54.48, 27.23, 18.43 ms.  Throughput: 216.19 iter/sec.
Timings for 6144K FFT length (12 cores, 8 workers): 55.81, 28.15, 56.01, 28.24, 55.96, 27.68, 54.97, 28.14 ms.  Throughput: 214.43 iter/sec.
Timings for 6144K FFT length (12 cores, 9 workers): 56.07, 55.81, 55.69, 57.16, 28.19, 54.79, 28.13, 55.04, 28.04 ms.  Throughput: 214.32 iter/sec.
Timings for 6144K FFT length (12 cores, 10 workers): 56.85, 56.18, 56.15, 56.48, 57.13, 56.48, 55.49, 27.64, 54.82, 28.61 ms.  Throughput: 213.50 iter/sec.
Timings for 6144K FFT length (12 cores, 11 workers): 56.95, 56.35, 56.35, 56.69, 57.15, 56.19, 56.71, 56.72, 56.71, 54.89, 27.73 ms.  Throughput: 213.17 iter/sec.
Timings for 6144K FFT length (12 cores, 12 workers): 57.10, 56.26, 56.27, 56.87, 57.22, 56.42, 55.97, 56.29, 55.89, 55.60, 55.90, 57.96 ms.  Throughput: 212.50 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 1 worker):  2.52 ms.  Throughput: 397.27 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 2 workers):  8.21,  8.21 ms.  Throughput: 243.66 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 3 workers): 12.20, 12.45, 15.77 ms.  Throughput: 225.72 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 4 workers): 18.32, 18.32, 18.22, 18.23 ms.  Throughput: 218.88 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 5 workers): 58.64, 27.88, 18.47, 18.29, 18.29 ms.  Throughput: 216.39 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 6 workers): 57.91, 28.29, 56.61, 28.59, 18.28, 18.30 ms.  Throughput: 214.62 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 7 workers): 57.49, 28.41, 56.79, 28.48, 58.42, 27.62, 18.43 ms.  Throughput: 212.90 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 8 workers): 58.06, 28.34, 56.11, 28.89, 56.78, 28.31, 57.00, 28.55 ms.  Throughput: 210.44 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 9 workers): 58.49, 56.56, 57.65, 56.73, 28.65, 58.29, 27.94, 55.94, 28.69 ms.  Throughput: 210.34 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 10 workers): 58.17, 56.90, 57.97, 56.52, 59.65, 56.51, 58.82, 27.83, 55.67, 28.87 ms.  Throughput: 209.70 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 11 workers): 57.10, 58.83, 57.91, 56.57, 58.55, 57.61, 56.74, 56.31, 58.41, 56.11, 28.96 ms.  Throughput: 208.75 iter/sec.
Timings for 6144K FFT length (12 cores hyperthreaded, 12 workers): 57.89, 56.77, 58.10, 56.50, 60.11, 56.28, 56.34, 56.34, 58.83, 55.91, 57.20, 59.63 ms.  Throughput: 208.84 iter/sec.

Assuming AMD's Zen 3 claims hold true, the next gen cpus will be >20% faster for both single and multithreaded workloads, given slightly higher clocks and that great IPC improvement. The 5800x being priced at 450 usd makes me sad though lol, at that point you might as well get the 12 core or save a bunch by getting the 6 core.

Nick · 2020-10-31, 13:48

If you watch a movie on a computer while you are overclocking it, does the movie run faster than it should?
If not, why not?

retina · 2020-10-31, 14:20

Quote:

Originally Posted by Nick

If you watch a movie on a computer while you are overclocking it, does the movie run faster than it should?
If not, why not?

If the computer you are asking about is your brain, then no. Overclock your brain and the movie appears to be slower.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Mandela effect, general discussion and how would you study it scientifically?	jasong	Lounge	12	2018-04-15 05:13
Is this solvable in general.	jwaltos	Homework Help	34	2017-09-21 14:12
S9 and general sieving discussion	Lennart	Conjectures 'R Us	31	2014-09-14 15:14
General Unix job-scheduling discussion	pinhodecarlos	Software	19	2014-04-17 15:04
General Status???	R.D. Silverman	NFSNET Discussion	4	2007-07-19 18:43

2020-10-06, 07:59	#14
Nick Dec 2012 The Netherlands 11001011100₂ Posts	OK well LaurV clearly doesn't think that that will be a quick way to get a deeper insight into the hardware, and he's an expert at this stuff! If you like hardware enough to consider reading a serious book on it, then the place to start is with the books of John Hennessy and David Patterson: "Computer Architecture" or "Computer Organization and Design". Your Dad says you like watching youtube: they gave a talk when they received the Turing award a couple of years back, and you can see that here: https://www.acm.org/hennessy-patterson-turing-lecture

2020-10-06, 22:47	#15
Xyzzy "Mike" Aug 2002 5·1,597 Posts	LaurV: Do you know of an easy way to crimp fan connectors without this super expensive tool? We have a pile of fans that need to have new tails put on them. (Finding the connectors and little crimp tabs isn't easy either.) See: http://tech.mattmillman.com/info/crimpconnectors/#kk100 Maybe the "PA-09" is okay?

2020-10-07, 00:01	#16
Uncwilly 6809 > 6502 """"""""""""""""""" Aug 2003 101×103 Posts 11×853 Posts	Why not buy 'adapters' or other similar leads (or find donor parts), cut and splice them. Use a bit of heat shrinkable tubing to protect from shorts.

2020-10-13, 15:27	#19
LaurV Romulan Interpreter Jun 2011 Thailand 5²×7×53 Posts	Hihi, I expected a comment on the fact that on the manometer, the green area is from 31 to 61, and the yellow is from 61 to 89 (and there are also other prime-related tricks on that screen, but maybe the image is not very clear).

2020-10-31, 13:48	#21
Nick Dec 2012 The Netherlands 2²·11·37 Posts	If you watch a movie on a computer while you are overclocking it, does the movie run faster than it should? If not, why not?