I bought a machine from them before with no difficulties, but it's hard to know for sure.
If there is fraud, I won't end up paying for anything
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
Sounds like an extortion racket. They put up a bunch of bad reviews and then the company has to pay to become a member before it can respond and have the bad reviews taken down. What a bunch of scum bags!
Regards,
Zenmastur
Only 2 defining forces have ever offered to die for you.....Jesus Christ and the American Soldier. One died for your soul, the other for your freedom.
I would like to keep my question in the same thread.
I have decided to buy a new computer to upgrade my chess analysis using LC0. If I buy 2 video cards (say RTX 2080) what processor would be reasonable to also run stockfish at the same time on a given position?
My idea is to run analysis on a move 15 Najdorf position and get opinions on the best move from both LC0 and SFish at the same time.
Would it be better to buy a 6-8 core with 2-2800 video cards for LC0 and then a second computer with 16-32 cores and a $100 video card for SF
OR
one massive investment that would have a 16-32 core, 2 -2080's and so on, to analyze both at the same time.
While money is a minor issue, I prefer the one BIG box over 2 "specialist" boxes. Advice? Pros? Cons?
Why go through the Bother and Hassle of buying and installing a whole new Setup of Motherboard, CPU, RAM and the like ?
The advent of NN Engines, which will only grow stronger with time, has made things easier, at least for me.
Just buy a couple of top-of-the-line RTX Cards, like I did, pop them into their Slots and you're ready to Match upto the strongest of CPUs and AB Engines.
Time and Ease of Installation are more important for me than money.
Even 64 Cores Users on Playchess are unable to beat me in LTCs, even though lc0 is relatively weaker than Stockfish.
My Hardware more than makes up for any weaknesses in Lc0, which will only improve with Time.
shrapnel wrote: ↑Mon Dec 09, 2019 4:40 pm
Even 64 Cores Users on Playchess are unable to beat me in LTCs ...
I love seeing "me" in your post. Are you doing anything at all? I mean, you're running an engine written by others, using a NN trained by others, on hardware that most users simply can't afford. Are you at least supplying an opening book that you've crafted yourself?
(You can tell I really don't understand the thrill you seem to get from this activity.)
Illusion wrote: ↑Mon Dec 09, 2019 4:03 pm
I would like to keep my question in the same thread.
I have decided to buy a new computer to upgrade my chess analysis using LC0. If I buy 2 video cards (say RTX 2080) what processor would be reasonable to also run stockfish at the same time on a given position?
My idea is to run analysis on a move 15 Najdorf position and get opinions on the best move from both LC0 and SFish at the same time.
Would it be better to buy a 6-8 core with 2-2800 video cards for LC0 and then a second computer with 16-32 cores and a $100 video card for SF
OR
one massive investment that would have a 16-32 core, 2 -2080's and so on, to analyze both at the same time.
While money is a minor issue, I prefer the one BIG box over 2 "specialist" boxes. Advice? Pros? Cons?
Thanks in advance.
A single box is better / cheaper.
Leela only needs 2-4 threads allocated.
Stockfish can run on the rest.
Not sure about your budget but you could wait for the 64 Threadripper that gets released early 2020.
shrapnel wrote: ↑Mon Dec 09, 2019 4:40 pm
Even 64 Cores Users on Playchess are unable to beat me in LTCs ...
I love seeing "me" in your post. Are you doing anything at all? I mean, you're running an engine written by others, using a NN trained by others, on hardware that most users simply can't afford. Are you at least supplying an opening book that you've crafted yourself?
(You can tell I really don't understand the thrill you seem to get from this activity.)
And what do you do with your dual xeon system pray tell?
Only 2 defining forces have ever offered to die for you.....Jesus Christ and the American Soldier. One died for your soul, the other for your freedom.
Illusion wrote: ↑Mon Dec 09, 2019 4:03 pm
I would like to keep my question in the same thread.
I have decided to buy a new computer to upgrade my chess analysis using LC0. If I buy 2 video cards (say RTX 2080) what processor would be reasonable to also run stockfish at the same time on a given position?
My idea is to run analysis on a move 15 Najdorf position and get opinions on the best move from both LC0 and SFish at the same time.
Would it be better to buy a 6-8 core with 2-2800 video cards for LC0 and then a second computer with 16-32 cores and a $100 video card for SF
OR
one massive investment that would have a 16-32 core, 2 -2080's and so on, to analyze both at the same time.
While money is a minor issue, I prefer the one BIG box over 2 "specialist" boxes. Advice? Pros? Cons?
Thanks in advance.
A nice cheap system would be a 3950X and dual 2070 supers. 28 threads for the AB program and the rest for Lc0. A more expensive system would be a 3970X and dual 2080Ti's. And an expensive system would be a 3990x and quad 2080Ti's.
Regards,
Zenmastur
Only 2 defining forces have ever offered to die for you.....Jesus Christ and the American Soldier. One died for your soul, the other for your freedom.
shrapnel wrote: ↑Mon Dec 09, 2019 4:40 pm
Even 64 Cores Users on Playchess are unable to beat me in LTCs ...
I love seeing "me" in your post. Are you doing anything at all? I mean, you're running an engine written by others, using a NN trained by others, on hardware that most users simply can't afford. Are you at least supplying an opening book that you've crafted yourself?
(You can tell I really don't understand the thrill you seem to get from this activity.)
And what do you do with your dual xeon system pray tell?
Research in diagrammatic/combinatorial knot theory and related graph theory and linear algebra over Z/2Z. And an occasional chess analysis too.
Are you doing anything at all? I mean, you're running an engine written by others, using a NN trained by others, on hardware that most users simply can't afford. Are you at least supplying an opening book that you've crafted yourself?