LCZero Accomplishments and Goals Thus Far

[frankp]

Where is the breakthrough in domain specific implementation exactly?

No one tells how to play - it learns for itself - perhaps ?

This is not domain specific as I understands it.
People tend to believe the alternative approach is somehow "superior", mislead by the results from a hardware overkill.

I was just pointing out that we seem to be comparing the merits of too very different approaches on the basis of the arrangements of semiconductors they use.

(It seems that any discussion of A0/leela turns into a A0/SF 'willy-waving' contest. I am still fascinated that leela play such a high standard of chess, not compared to SF of course, on my consumers grade two generation old graphics card.)

[Albert Silver]

Sigh..wake me up when it is 2800 elo running on singe CPU core, which is what every other engine uses in rating lists. As far as I am concerned, it is still a 2100 elo engine there.
I see so many excited people giving a hardware advantage to LCzero, like CCLS does for instance uses a GPU for LCzero and single core CPU for the rest of the engines.

I don't think anybody misses the fact that given a hardware advantage that will basically make the evaluation free, you can increase your elo to your satisfaction.
If Stockfish had its evaluation FGPA'ed it may be a 4000 elo engine but who cares for that anyway ? The fact is Stockfish has been throwing away evaluation features for the sake of speed throughout the years.

Daniel

I wonder 2100 at what time control and if the rating is not higher at longer time control.

Edit:I am not sure if the part that stockfish throwed away were productive for stockfish even with no price in speed and I guess that stockfish had counter productive code in the evaluation not because it made it slower but because something that humans believe to help simply does not help.

I have an i5-2500K (quad-core) and a GTX1060. Playing g/10 matches, Leela 202 has had a ~2900 CCRL performance. The opponents used the quad. Note also that Leela does NOT use 100% of the CPUs, while the opponents do. This is seen clearly in the task manager.

[noobpwnftw]

You are comparing TFLOPS, which is irrelevant.

.

But any 'hardware avantage' should be measured in $$$, (possibly including the price of electricity);, other ways just don't make any sense.

I only compare GPU to GPU, to show how a TPU is different than a ordinary GPU which you can get with a few hundred bucks.

There is a limited number of GPUs you can pack in a "single machine", so does CPUs, so the per unit performance matters, no? Without that, how do you compare the price tag?

[hgm]

Exactly. The whole misconception that AlphaZero had any hardware advantage at all comes from ridiculously overweighting the importance of multiplication, because the CPU is extremely poor at that, and Stockfish happens to not need it. It is like claiming a wheelbarrow has an extreme hardware advantage over a helicopter, by only counting the number of wheels.

[noobpwnftw]

I was just pointing out that we seem to be comparing the merits of too very different approaches on the basis of the arrangements of semiconductors they use.

(It seems that any discussion of A0/leela turns into a A0/SF 'willy-waving' contest. I am still fascinated that leela play such a high standard of chess, not compared to SF of course, on my consumers grade two generation old graphics card.)

I think there is nothing wrong with the approach, and it appears to scale well because we are at the stage where people usually have 1 GPU in their computers, which is easier to do doubling than where they already had some 8 CPU cores.

[hgm]

There is a limited number of GPUs you can pack in a "single machine", so does CPUs, so the per unit performance matters, no? Without that, how do you compare the price tag?

The four TPUs in AlphaZero's hardware were just a single card fitting in a slot of the PCI bus of the PC they were running it on. So I don't see what point you are trying to make.

[JJJ]

Is autoresign really a good Idea ? At fast time control I see Leela switching a lot between winning and loosing. Leela still need to play better in ending instead of resigning too fast.

[noobpwnftw]

There is a limited number of GPUs you can pack in a "single machine", so does CPUs, so the per unit performance matters, no? Without that, how do you compare the price tag?

The four TPUs in AlphaZero's hardware were just a single card fitting in a slot of the PCI bus of the PC they were running it on. So I don't see what point you are trying to make.

From the current implementation(A0, Leela) it appears that the input pre-processing is done on the CPU-side, which means the IO throughput of a PCI-E bus is also a factor.

CPUs only have limited lanes for PCI-E which limits how many GPUs a program can drive in parallel efficiently.

Theoretically you can use layered bus extenders to link several hundred GPUs, but then you get poor IO performance like the mining rigs.
They even use Celeron CPUs to drive a good number of modern GPUs, because their programs are mostly autonomous on the GPU-side, which is not the case here.

When it comes to scalability, those things do matter, let's just say you have a machine that has a max number of GPUs, probably due to the above limiting factor, how to go beyond that and still "scale well" is a question.

[Daniel Shawul]

Sigh..wake me up when it is 2800 elo running on singe CPU core, which is what every other engine uses in rating lists. As far as I am concerned, it is still a 2100 elo engine there.

Well, on my hw Leela seems to use CPU 2 cores + GPU, so I played a short match against my engine and Leela performed like 2800+ engine already (it was net 217) and the TC was 40/1 minute so rather fast for Leela.
Also people who want a strong chess entity don't care about 1 CPU.

LCZero can run on the CPU so I've made comparisons with ID 125 using 2 threads and got about 2100 elo on an i7 laptop. If I used the Integrated intel HD graphics card I have instead, the result would be worse I am sure as it has slower nps there.

TCEC made a fair 44-core comparsions that probably showed it scales better than other engines, but the fact remains that you have to use either massive hardware or a 1 year + 1 month time control for it on single CPU core.

[frankp]

I was just pointing out that we seem to be comparing the merits of too very different approaches on the basis of the arrangements of semiconductors they use.

(It seems that any discussion of A0/leela turns into a A0/SF 'willy-waving' contest. I am still fascinated that leela play such a high standard of chess, not compared to SF of course, on my consumers grade two generation old graphics card.)

I think there is nothing wrong with the approach, and it appears to scale well because we are at the stage where people usually have 1 GPU in their computers, which is easier to do doubling than where they already had some 8 CPU cores.

Yes, from a practical point of view, I guess both the ability to train leela and its strength in use (NN size and speed) will be limited, perhaps rather quickly, by the power of affordable consumer grade graphics cards. Well at least in the short term. Still interesting to see how far it can go. Still a young project.