Posts that seem less than useless, or something like that

[davieddy]

Originally Posted by TheMawn

I think Trial Factoring will get you the most GHz-Days and we aren't by any stretch too far ahead of the LL-front to be adding more firepower to TF.

Now that you are a veteran of 4 months and 300 posts (not to mention the odd dig at me) I think we can declare the honeymoon to be over and the gloves off.

1) Comparing TF on GPU with LL on CPU is more ridiculous than apples and oranges.

2) The second part is simply blatantly self-contradictory.

3) If you are sick of me beating up Chalsall with 3 arms tied behind my back, just consider how we might feel about it. Until one of us FFFades away, we are obliged to. I suppose he may will see the light soon enough.

Keep on truckin'.

[davieddy]

Originally Posted by davieddy

folk other than you, Chris and I may not be quite up to speed on this thread. Rather like TF.

I was about to post this in the Musak thread, but this time and place seems suddenly appropriate.

Did anyone else first hear Far Away Eyes on a pub jukebox, near the end of a long, beautiful, even if painful day out with a beloved ex-girlfriend?

David

Originally Posted by davieddy

I'll take this as token of appreciation.
Of course the circumstances of my first hearing are mercifully rare.

Did you notice 5 beats to at least some of the bars/gin-joints?

If I suggested they must have been listening to Sweetheart of the Rodeo I don't think I would be 100 years away.

Love as always,
D

Originally Posted by kladner

Um, I don't quite follow that.....

I wanted "Miss You", but we have since become extremely fond of "Faraway Eyes."

Ain't it funny how the standard of moderation has plummeted recently?
Kieran and I both acknowledged that our musical offerings could have been posted here, but were in their way better in tthe context. I suppose we should be thankful that Chalsall hasn't emasculated them as of yet (touch wood).

I haven't quite sussed out the beat pattern on Faraway eyes yet either. This might well be part of its mesmerizing charm. It is certainly not as blatant as Dave Brubeck's "Take Five" or even Pink Floyds "Money".

Often when you actually suss the simple reason a track captivates you, it partially spoils your naive enjoyment of it forever.

Without wishing to spoil anyone's enjoyment of this one
(that is impossible) it doesn't take too much listening to suss what exactly ties your heartstrings in knots.

D

[davieddy]

I once bought an old electronic organ, and each note had its own tunable oscillator.

I tuned the frequncies of the major scale in the Pythagorean proportions of 8, 9, 10, 32/3, 12, 40/3, 14.

The major chord sounded great. Adding the 7th (and 9th and 10th) was a revelation, even though being I violinist I knew what an interval of 7/6, 8/7 and 9/8 could sound like.
The striking difference from equallly tempered scale (semitone = 2^1/12 is that in the key of C, 14/8 is roughly midway between A and B flat.

I can assure whichever supermod has the thankless task of scrutinizng my posts that this will be comprehensible/of interest to at least one regular here.

Please afford me the opportunoty of editing, self-censorship and above all spontaneity. I might then regain the SOH on which I used to pride myself.

David

[davieddy]

Originally Posted by kladner

Heh! As a degree-less person I find those quite amusing.

Surely you must have encountered thethird degree in your time?

[davieddy]

Originally Posted by TheMawn

Why do we need a four month buffer (50,000 candidates at ~400 LL's per day)?

Ome on man.
Read the explanation the resof the paragraph.
Virtually nothing is lost if we do have this buffer, and it affords more time for P-1.

But the important reason is that TF from 73 to 74 may be keeping pace with LL assignment (although I have a strong hunch that LL assigment is being held up. It can't be good to assign 66M exponents while 62M ones are being hoarded for futile TF to 74), but TF from 71 to 74 will take almost twice as long as 73 to 74. With a decent buffer, I have estimated that the TF rate might match (and then surpass) the LL assignment rate before assinment has caught up, and we get into the disgraceful mess we are currently in. I have been saying this ever since Chris go to greedy TF-wise.
TF is essential, but should never get near interfering with LL assignment.

You know I am so sure of my ground that I will not change my stance.
I have studied this topic more than anyone since mfaktc was created.
It is regrettable that Pete ceased his obviously valuable contribution, but if you read his posts which led up to it, he simply ridiculed me (with no small encouragement from Chris) instead attempting to understand my point or even request clarification. Except in a sneering rhetorical "questions".

I have received two bans for standing my ubrefuted ground.
Chalsall offered the most pathetic excuse for the first one - "Pete is more important than you are" as if it were a case of him or me. Maybe I might have persaded Pete that his firepower was valued and suggest he stuck at it.
Serge's swinging 2 month ban was imposed simply for reiterating my consistent in slightly powerful language.
Believ me, if you are sick of me arguing with Chris, just spare a thought for me.

What is going to happen when the supply of expos TFed to 73 dries up?
Chris has said that he automatically going to start assigning expos TFed to 73. Oops, there will soon be no such exponents.

If this carefully composed post doesn't get published, I shall despair.

David

PS If someone can refute any of this, I couldn't have put it more clearly.
Any criticism needs to be carefully considered.

[davieddy]

Originally Posted by davieddy

Did anyone else first hear Far Away Eyes on a pub jukebox, near the end of a long, beautiful, even if painful day out with a beloved ex-girlfriend?

Originally Posted by kladner

I first heard it on the flip side of the 12" extended single of "Miss You", which I stole from my next younger brother decades ago.

Originally Posted by davieddy

I'll take this as token of appreciation.
Of course the circumstances of my first hearing are mercifully rare.

Did you notice 5 beats to at least some of the bars/gin-joints?

If I suggested they must have been listening to Sweetheart of the Rodeo I don't think I would be 100 years away.

Originally Posted by kladner

Um, I don't quite follow that.....

Neither do I anymore:
It seems to be simply 4 time.

My confusion arose because I was trying to sing "Hickory Wind " or "The Christian Life" to it (or vice versa) .
Now Brahms was a master of combining 3 and 4 beats in the same bar.

BTW another connection between the two songs is that Gram was a great mate of Keith and was staying with them in the south of France around the time of "Exile on Main Street". Cue Rip This Joint

D




D

[davieddy]

Originally Posted by chalsall

David, you come across a bit like a theoretical physicist who never considers the empirical evidence (or, even worse, rejects it because it doesn't fit the theory)...

How does this tally in the slightest with this self-description of my credentials?
However, the tone your typical, unconsidered remark tallies perfectly with my (admittedly cutting and ever more bitter) criticism of you.
Why should you expect (as you seem to) to "beat" or even effectively criticise someone 15 years your elder without putting in at least a modicum of consideration?

BTW I do understand why many here are getting sick to death of our arguments, and so am I. But if you insist on appointing youself judge, jury and executioner, we appear to be condemned to eternal wrangling.

I hope you will read my latest post to George very carefully, I would recommend discussing its content^TM with him before putting your foot in it yet again.

Originally Posted by davieddy

...many opt for "What Makes Sense". Now Chalsall is not sensible, and is anyway biassed by his wish to promote his GPUto72 mob's achievements and potential at all costs.There are of course several people who could point out any shortcomings but as you noted in your first sentence, my determination to get my simple TF stategy across is thwarted by witholding my posts until they are out of date, and Chalsall's near fanatical determination to do the opposite of what I suggest, and humiliate me with all means at his disposal which are far too many, and his conduct is unbecoming of a supermod.
As a very assiduous and objective observer of the project and a first class physicist of 60 years experience, no-one should dispute my conclusions without very careful consideration. This is where Chris and I diverge the greatest.
I can spot how he plays fast and loose with statistics, and as I am sure some of you are aware, this practice is anathema to any physicist worth his salt.

Originally Posted by davieddy

Indeed. His tables are obligatory reading for me.
Why won't he acknowledge that this page is blank for 12 hours a day, and apply the trivial fix?

[davieddy]

Originally Posted by chalsall

no LLer will ever be given a candidate above 60M TFed to less than 73,

I accept this.

and above 63M any less than 74.

AND PIGS MIGHT FLY.

[davieddy]

Originally Posted by TheMawn

I think we get one hour to review our posts which ought to be more than enough, but there's something to be said for sleeping on a post.

How about a mandatory reason-for-editing? I like the idea of a post history also.

I fully accept your apology, and am grateful for your thoughtfulness and undrstanding.
I hope whoever has failed to lift the unbearable delays in publishing my posts will see that this often causes embarrassment to many besides me.

PM is sometimes helpful, but the B******s have limited my post box to 11, and a maximum of one recipient at a time.

Chalsall even posted publicly that it was his intention to make me angry. What a charming man. Any wonder that he sometimes succeeds?

What was that someone said about bullying?

D

[davieddy]

Originally Posted by chalsall

If it's profitable to TF to 73.5, for example, then ideally we would TF half of the candidates to 73, and the other half to 74. (It's a little more complicated than that, but hopefully you get the idea.)

If it's profitable to TF to 74.0 or above, then ideally we would TF all candidates to at least 74.

This isn't Rocket Science, nor Brain Surgery....

You haven't read my post to George have you?

You kow perfectly well that we are limited to what is feasible.
If we have two exponents TFed 73, and can take one but not both to 74,
should it be the higher or lower one? There is no contest.
Now this is definitely not Rocket Science.
Brain Surgery is what you need.

[davieddy]

Originally Posted by Prime95

No, it means that each candidate would be TF'ed halfway between 2^73 and 2^74 (i.e 2^73.5). This would require a change to mfaktc which only supports TFing to 2^73 or 2^74, not 2^73.5. It would also require a change to Primenet to allow reporting "no factor to 2^73.5".

The good news is that for a candidate where the optimal amount of TF is 73.5, the inefficiency introduced by TFing to 2^73 or 2^74 is mighty small.

I understand that mfaktc allows you to TF to 73.5, but the simplicity of sticking to integer bit levels is something we are all taking for granted in this discussion mudfight.
However since the time difference between 73 and 74 for a given exponent is a factor of ?2, it makes sense get the decision right, i.e trust me who has had to endue a year's cruelty for the cause, because he can and does evaluate the evidence expertly and objectively, and not that self-publicist Chris who, for obvious reasons, wears rose-tinted glasses and the emperor's new clothes.

Originally Posted by Prime95

This is a good assumption for our theoretical discussions here.

Well how good it is, is an imponderable question about the future, but I had hoped that its simplicity would enable you and others to follow the rest of my argument, especially since it involved nothing more nor less than a simple refutation of a simple helpful suggestion from Philmoore.
The wiggle room remaining is "what is x?" which remains to be evaluated from the data available. Deciding this is obfuscated by Chalsall's willingness to allow TF to interfere with LL assignment which rightly worries many of us, and I maintain is limiting the assignment rate
to the TF rate. This is a simply testable hypothesis - simply release the few remaining expos TFed to 73 and see what happens. It will also enable us to decide whether TF from 71 to 74. Chris claims to have an empirical
aproach to these questions and likes to say I don't.
I don't say "I am a physicist" for nothing. Being a competent theorist does no harm, and having theories tested thoroughly is what we want, and suggesting suitable experiments to do this is part of our job.
All that remains then is for Chris to "Suck it and See". I feel it he who is scared of what might happen. .

However, should chalsall's TF GPU'ers fall seriously behind, I have the option of diverting some or all of prime95's "do what makes the most sense" users from LL to double-checking.

Actually, I'm considering this a little bit anyway as LL'ers are pulling way ahead of the double-checkers.

Well that is a silly cop out and a different matter.
It will certainly decrease the pobability of finding a prime within, say, two years, and DCs completed 3 times slower than LLs makes good sense, an assertion I will explain another time.

As I said, TFing won't "fall behind" LL if you allow it to hold back LL assignment in the way it is ATM.

Come of the fence George, PLEASE. And untie my hands.

David