Sandy Bridge - Bulldozer - Computer Chess Test!

[vmsein]

Here is my latest run:


Thanks Jean-Paul!

[JPQY]

The Watcher...vmsein...SteveRo & freshy...Thanks for your results

List updated!

[T_Flight]

I have been trying to figure out what is causing these things to drop off as they progress, and for the life of me cannot figure it out. They start off at 1-2 seconds with a bang, but the nodes/s drop off...especially towards the tail end of the run.

Let's hear ideas on this one. I'm all ears.

[JPQY]

Hi T_Flight...yes you are right,there is a drop in speed.
I try to explain why! This is not a bench program,like most people think,it's a full working chess program with a high level engine.
Every programmer gives to there engine alot settings...to make it not complicated i have put these settings in advance in the program,so that it's
for every body same testing with same settings.

On every screen you see two values memory(i will take the screen here on this page above from vmsein as example)
1.Up in the middle you see 4096Mb = is his free memory the system have.
2.Under in Analyse window,just next the name TogaII142JD-8cpu you see 135Mb = Hash Tables = HT

What is HT = i have give the engine 135Mb HT is memory he use where he can put his calculatings he wants to keep!
every time he goes one Depth deeper,he can look in this memory(data) so that he has not to calculate the same line anymore that he already knows.
This goes on till this memory is full !! from that moment you get a drop in speed,because when he calculate further,find better lines,he has to remove
old data to make place for new better data!

Ofcourse you can give more HT to the engine so long you have free memory...if you have 6Gb...you can give 4Gb HT to the engine if you want...and the engine
can keep alot more data in his memory! and will go a little faster..till memory is full.

This is just a little test,because most people get use to test there system with known Benchmarks..like Pi..just a few seconds,and done..but with chess is it endless
you can continue hours,days,forever calculating till chess is calculate out!
So i have to make something small..that don't take long time,and acceptable for most testers.

I hope this simple explain is clear enough..because chess is much more then just caculatings!

Oh yes..chess is the most stressing game you can find,and we as chess players,don't like so much Vista,because it slow down the system,you get lower
calculating speed,so we stay with XP...some engines get even a gain from 60% speed if you run Win XP 64bit!!
But that's only for Xtreme people who want the max. performance!

JP.

[dejanh]

Time to add my name to the book as well

Processor is Core i7 940, overclock is 23x179MHz (4117MHz), 4 cores/8 threads, turbo on at 1.46250V BIOS (1.493V measured at load), all other settings auto. Memory is at 1794MHz with DRAM voltage of 1.67006V BIOS (1.652V measured at load), 8-8-8-24-2T timings.

Screenshot and the results file is attached.



Got another set of results. These are a bit better I have attached them as chess_test_2.txt.

[Aberration]

Here's what I got.

[T_Flight]

Nice! That bench puts the heat on doesn't it? hee heeee

I noticed your numbers seem to be a bit mroe stable than mine. It might be due to your watercooling. I suspect becasue it's running 10 dgrees cooling than mine at around the same clocks that your's is more efficient. Mine runs 75 around those clocks.

I'm probably gonna go ahead and do a few more with what I have until I get my water loop up. I've gotta get it going until I can do some horse trading on that RAM. It's stiull scattered over the card table and will eb until I get my water loop up and tested, and then we'll get to the pretty stuff.

I'm getting ready to play with some clocks now.

[Aberration]

LinX gets about 7-8C higher.

[JPQY]

dejanh & Aberration...Thanks for your results

List updated!

[JPQY]

Kensek..thanks..you have taken the 3 place with a retail core i7 965 @4.320Ghz!!

[speckled10]

Here is a trial run @ 4.1Ghz ..



21/48 02:17 139.982.059 8.139.615 +0.25 e2e4 e7e5 Ng1f3 Ng8f6 d2d4 e5xd4 e4e5 Qd8e7 Bf1e2 Nf6g4 Qd1xd4 h7h5 Qd4f4 Nb8c6 Nb1c3 Nc6xe5 00 c7c6 h2h3 Ne5xf3+ Be2xf3 Ng4e5 Bf3e2 d7d5 Bc1e3 Qe7f6 Qf4xf6 g7xf6
21/47 04:20 259.464.922 8.217.990 +0.30 e2e4 e7e5 Ng1f3 Ng8f6 d2d4 e5xd4 e4e5 Qd8e7 Bf1d3 d7d6 00 d6xe5 Nf3xe5 Nb8d7 Ne5f3 Nd7c5 Rf1e1 Bc8e6 Nf3xd4 Nc5xd3 c2xd3 000 Nd4xe6 f7xe6 Nb1c3 Kc8b8

Nice program !!

EDIT: I notice StrangerS is significantly faster in benchmark (8.6xx.xxx) with virtually identical setup to mine - but he is running XP .. is program optimised for XP or Vista (64-bit) ?

[JPQY]

Thanks speckled10 for your result and perfect screen

If you look to my second message here on page 4 , I have explain a little that Vista slow down the speed from the cpu,
because you have many background programs running..and chess program needs/wants 100% cpu! So we stay with XP!
Hope Windows 7 will be better!

JP.

[JPQY]

Today Kensek comes very..very close to the first place!!

1) TogaII142JD-8cpu 21/44 01:52 122.655.156 8.747.857 +0,22 Nehalem @4.347Ghz(Core i7 965ES)DDR3: 8-8-8-20 Rol-co
2) TogaII142JD-8cpu 21/48 02:00 131.600.346 8.736.667 +0.25 Core i7 965 @4.284Ghz DDR3: 9-9-9-24 Kensek New

JP.

[JPQY]

And we have a new leader...Thanks to Kensek!!

1) TogaII142JD-8cpu 21/49 02:19 152.759.022 8.761.006 +0.22 Core i7 965 @4.318Ghz DDR3: 9-9-9-20 Kensek New

JP.

[speckled10]

If you look to my second message here on page 4 , I have explain a little that Vista slow down the speed from the cpu,
because you have many background programs running..and chess program needs/wants 100% cpu! So we stay with XP!
Hope Windows 7 will be better!

JP.

Thanks JP ..

I am a little shocked that Vista chews up so many resources with background tasks ?? These must be invisible to the user because the task manager shows no other activity while your program is running ..

Is Vista really that much of a burden on CPU over XP ? Does it affect all programs the same way ?

If this is the case - I think it would be good to list which OS (Vista or XP, 32-bit or 64-bit) people ran their tests on, to make it a fair comparison to all ..

[Metroid]

Hi JP,

Well since I had upgraded my system with watercooling. I felt that I needed to make some more tests.





This is the best I could do with my limited ram Crucial 1066Mhz 7.7.7.21. I got an astonishing 1600Mhz 9.9.9.24 1.64v. If I had a better ram then I would beat all the others scores no problem and get the first place, 8.800 would be the minimum.

Metroid [29.12.2008] [21.45] [8.537] [I7 920 4.2Ghz 1.40V 21x200 P6T Deluxe 1600Mhz 9.9.9.24 Win XP] Arena+TogaII142JD-8CPU

[BulldogPO]

Here:


And some others:
Deep Shredder 11 UCI 64bit:
http://bulldogpo.wippiespace.com/i7_...dder_Chess.png

Fritz Chessbench 9:
http://bulldogpo.wippiespace.com/i7_...ritz_Chess.png

[JPQY]

Metroid & BulldogP0...Thanks for your results

List updated!

[rge]

Run at 4.4...
Edit..1 more run at 4.43...all tinkering I can do today...heading to ny with family for new year.
3x1gb ram (xpsp3), UC 18x (same mhz as qpi)
Edit 2: second run at faster mhz was slower...guess playing with email at same time was bad idea....will rerun when back in town in week

[Rol-Co]

21/47 02:23 123.766.593 10.946.018 +0,19 e2e4 e7e5 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 d2d3 Ne4f6 d3d4 Bf8e7 Nb1c3 00 Bf1b5 Bc8d7 Bb5xd7 Qd8xd7 00 Nb8c6 Rf1e1 Qd7g4 Bc1g5
21/49 02:24 118.483.942 10.917.105 +0,20 e2e4 e7e5 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 d2d3 Ne4f6 d3d4 Bf8e7 Nb1c3 00 Bf1b5 Bc8d7 Bb5xd7 Qd8xd7 00 Nb8c6 Rf1e1 Rf8e8 Bc1f4
21/51 02:24 172.890.406 10.882.316 +0,23 e2e4 e7e5 Ng1f3 Ng8f6 d2d4 Nf6xe4 Bf1d3 d7d5 Nf3xe5 Bf8d6 00 00 Nb1c3 f7f5 Bc1f4 Nb8c6 Bd3xe4 d5xe4 Ne5xc6 b7xc6 Bf4xd6 c7xd6 Qd1h5 Bc8e6

[Kensek]

This chess program prefers low BClock and a High Multi.

125 x 35 does way better than 208 x 21.

FEN: rnbqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/RNBQKBNR b KQkq - 0 1

TogaII142JD-8cpu:
1/1 00:00 2 0 +0.06 Nb1a3
1/1 00:00 3 0 +0.34 Nb1c3
1/1 00:00 13 0 +0.40 d2d4
2/2 00:00 45 0 +0.20 d2d4 d7d5
3/8 00:00 205 0 +0.34 d2d4 d7d5 Ng1f3
4/8 00:00 565 0 +0.20 d2d4 d7d5 Ng1f3 Ng8f6
5/12 00:00 2.300 0 +0.30 d2d4 d7d5 Ng1f3 Ng8f6 Nb1c3
6/12 00:00 3.761 0 +0.20 d2d4 d7d5 Ng1f3 Ng8f6 Nb1c3 Nb8c6
7/14 00:00 10.306 0 +0.20 d2d4 d7d5 Ng1f3 Ng8f6 Nb1c3 Nb8c6 Bc1f4
8/14 00:00 21.704 0 +0.20 d2d4 d7d5 Ng1f3 Ng8f6 Nb1c3 Nb8c6 Bc1f4 Bc8f5
9/16 00:00 33.902 0 +0.07 d2d4 d7d5 Ng1f3 Ng8f6 Nb1c3 Nb8c6 Qd1d3 Bc8g4 Bc1f4
9/19 00:00 48.807 0 +0.14 Ng1f3 d7d5 Nb1c3 Nb8c6 e2e4 d5xe4 Nc3xe4 Bc8f5 d2d3 Ng8f6
10/24 00:00 85.104 0 +0.23 d2d4 d7d5 Ng1f3 Ng8f6 Nb1c3 Qd8d6 Bc1d2 Nb8c6
10/22 00:00 88.306 0 +0.24 Ng1f3 d7d5 Nb1c3 Nb8c6 d2d4 Ng8f6 Bc1f4 b7b6 h2h4
10/24 00:00 87.098 0 +0.26 d2d4 d7d5 Nb1c3 Ng8f6 Ng1f3 Nb8c6 Bc1f4 Bc8f5 Nc3b5 Ra8c8
11/24 00:00 121.733 0 +0.13 d2d4 d7d5 Nb1c3 Ng8f6 Ng1f3 Nb8c6 a2a3 Bc8f5 Bc1f4 e7e6
11/22 00:00 108.293 0 +0.16 Nb1c3 d7d5 d2d4 Ng8f6 Ng1f3 Nb8c6 Bc1f4 Bc8f5 h2h4 e7e6
11/25 00:00 150.429 0 +0.20 e2e4 d7d5 e4xd5 Ng8f6 Ng1f3 Nf6xd5 Bf1c4 Nd5b6 Bc4b5+ c7c6 Bb5e2 Nb6d5 00 Bc8f5
11/23 00:00 189.621 0 +0.24 e2e4 d7d5 e4xd5 Ng8f6 Ng1f3 Nf6xd5 Bf1c4 Nb8c6 00 Nc6a5 Bc4xd5 Qd8xd5 d2d4
11/24 00:00 202.687 0 +0.28 e2e4 Ng8f6 e4e5 Nf6d5 c2c4 Nd5b6 Nb1c3 d7d6 Ng1f3 Nb8c6
12/25 00:00 306.246 0 +0.28 e2e4 d7d5 e4xd5 Ng8f6 Ng1f3 Nf6xd5 d2d4 Bc8f5 Bf1b5+ Nb8d7 00 c7c6 Bb5d3 Bf5xd3 Qd1xd3 e7e6
12/24 00:00 316.609 0 +0.31 e2e4 d7d5 e4xd5 Ng8f6 Ng1f3 Nf6xd5 Bf1c4 Nd5b6 Bc4e2 Nb8c6 00 Bc8f5 d2d4
13/31 00:00 533.453 0 +0.32 e2e4 d7d5 e4xd5 Ng8f6 Ng1f3 Nf6xd5 d2d4 Bc8f5 Bf1b5+ c7c6 Bb5d3 Bf5xd3 Qd1xd3 Nd5b4 Qd3b3 Nb8a6
14/30 00:01 1.321.957 9.280.000 +0.32 e2e4 d7d5 e4xd5 Ng8f6 Ng1f3 Nf6xd5 Bf1c4 e7e6 d2d4 Nb8c6 00 Bf8e7 Nb1c3 00 Rf1e1 Nd5xc3 b2xc3
14/34 00:01 1.669.962 9.108.480 +0.34 e2e4 e7e6 Ng1f3 d7d5 e4xd5 e6xd5 d2d4 Ng8f6 Bf1b5+ Bc8d7 Bb5xd7+ Nb8xd7 00 Bf8b4 a2a3 Bb4d6 Rf1e1+ Nf6e4 Nb1c3
14/32 00:01 1.685.992 9.108.480 +0.35 e2e4 e7e6 Ng1f3 d7d5 e4e5 Bf8e7 d2d4 c7c5 Nb1c3 Bc8d7 a2a3 c5xd4 Nf3xd4
15/34 00:02 2.378.503 9.040.000 +0.34 e2e4 e7e6 Ng1f3 d7d5 e4e5 Ng8h6 d2d4 Nh6f5 g2g4 Nf5h4 Nb1c3 Bf8b4 Bf1d3 00
16/39 00:04 4.552.948 8.800.000 +0.33 e2e4 d7d5 e4xd5 Ng8f6 d2d4 Nf6xd5 Ng1f3 e7e6 Bf1c4 Bf8b4+ Bc1d2 Nb8c6 00 00 Nb1c3 Bb4e7 Bc4d3 Nd5xc3 Bd2xc3
17/39 00:07 8.898.137 8.630.000 +0.19 e2e4 e7e5 Ng1f3 Nb8c6 Bf1b5 Bf8c5 Nb1c3 Ng8f6 d2d3 00 00 d7d6 Bc1e3 Bc5xe3 f2xe3 Bc8d7 Bb5xc6 Bd7xc6
17/36 00:09 10.748.610 8.640.000 +0.26 e2e4 e7e5 Ng1f3 Nb8c6 Bf1b5 Bf8c5 Nb1c3 Ng8f6 d2d3 00 00 d7d6 Bc1e3 Bc5xe3 f2xe3 Bc8d7 Nf3g5 Kg8h8 Bb5xc6 Bd7xc6
17/39 00:13 15.165.652 8.646.154 +0.37 e2e4 e7e5 Ng1f3 Nb8c6 Bf1b5 a7a6 Bb5xc6 d7xc6 00 Bc8g4 h2h3 Bg4xf3 Qd1xf3 Ng8f6 d2d3 Bf8d6 Nb1c3 00
18/40 00:15 16.799.142 8.677.333 +0.26 e2e4 e7e5 Ng1f3 Nb8c6 Bf1b5 Bf8c5 Nb1c3 Ng8f6 d2d3 00 00 d7d6 Bc1e3 Bc5xe3 f2xe3 Bc8d7 Bb5xc6 Bd7xc6 Nf3g5 Kg8h8
19/44 00:31 35.075.000 8.784.516 +0.26 e2e4 e7e5 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1e3 Bc8e6 000 000 Qe2b5 a7a6 Qb5a4 d6d5
20/44 00:58 64.973.843 8.824.828 +0.21 e2e4 e7e5 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1f4 Bc8f5 000 Qe7xe2 Bf1xe2 Bf8e7 Nf3d4 Nc6xd4 Rd1xd4 Ke8d7 Rh1e1
20/45 00:58 64.989.643 8.824.828 +0.23 e2e4 e7e5 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1f4 Qe7xe2+ Bf1xe2 Bf8e7 Nf3d4 Nc6xd4 c3xd4 Bc8f5 Ke1d2 00 Rh1e1
20/44 00:58 64.648.653 8.824.828 +0.24 e2e4 e7e5 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1f4 Qe7xe2+ Bf1xe2 Bf8e7 Nf3d4 Nc6xd4 c3xd4 Bc8f5 000 000 Rd1e1
21/50 02:25 160.854.229 8.840.828 +0.19 e2e4 e7e5 Ng1f3 Ng8f6 d2d4 Nf6xe4 Bf1d3 d7d5 Nf3xe5 Bf8d6 c2c4 Bd6xe5 d4xe5 Nb8c6 00 00 Bd3c2 Nc6xe5 Qd1xd5 Qd8xd5 c4xd5 Ne4c5 Bc1f4 Rf8e8 Nb1c3
21/50 02:25 159.421.966 8.840.828 +0.22 e2e4 e7e5 Ng1f3 Ng8f6 d2d4 e5xd4 e4e5 Nf6e4 Qd1xd4 f7f5 e5xf6/ep Ne4xf6 Bf1d3 Nb8c6 Qd4h4 Qd8e7+ Ke1d1 d7d5 Bd3g6+ h7xg6 Qh4xh8 Bc8e6 Bc1e3 000 Nb1c3

[T_Flight]

Shhhhh, that's a secret. We wouldn't want to actually say that it has a practical use.

Seriosuly though. There are uses for these Extreme Edition CPU's. That's what I really like about this Chess program. It really Bullies the CPU, and these calcs show deficiencies that allow you to tune your system for real world stuff...not just a number. I have seen first hand that while tuning my system for this app, that other apps responded also. Things like simulation runs on my rocket motor software improved dramatically.

Setting this thing to Infinity and letting it rip will really show a user what efficiency really is, and if there are any bottlenecks or settings not quite right it will show it. Soon I hope to get mine into the 8,000,000 range. I've got a 2 week wait until I get get all my watercooling stuff here finally. I should be able to take it into the 4-4.2 range. Maybe I'll luck out and be able to get to 4.5 if this CPU is golden.

[JPQY]

Hey all..first i want to say to everybody a nice Happy New Year and a good health for 2009!!

I have tried a few time to let my little test running,that people will see a different way how you can tune your system with a daily use program!
And this time it goes very well...and i can't say better what T_Flight just write..my English is not good enough to explain everything what i want
to say from this program..i let it see or run as a bench,but it's much more..you are busy with professional chess software who has many chess knowledge.
The chess grandmasters getting defeat by these chess software from today..so strong they are,but still no computer can caculate chess out!
Here i let just computers calculate around 2-3min. but for me it can be days busy to find one problem..it's endless for the moment...but already amazing
what they can find,because computers comes faster and faster but also the chess knowledge put in the program is fantastic!!
Thanks to many years programming to make these software stronger-better and faster then ever!

Now..i give my thanks word to all testers who try my test and like to see the fighting spirit from some people who wants always better!

Mr. Kensek has taken at last moment the first place so he is the number 1 from 2008!!
But the battle is not finished yet..i know some persons who want to take back this first place,so i think 2009 will be interesting again,
and also with the comming Nehalem EP or dual core i7 socket we will see new high scores.

And don't worry..there can be never enough cores for chess!

JP.

[bowman]

Got a new one for you..

http://www.isarapix.org/pix95/1230866159.png

It might not have a multiplier in the 30's but that's not a privilege I'm gonna pay €800 for.

FEN: rnbqkbnr/pppp1ppp/8/4p3/4P3/5N2/PPPP1PPP/RNBQKB1R b KQkq e6 1 2

TogaII142JD-8cpu:
1/1 00:00 2 0 +0,30 Ng1f3
1/3 00:00 6 0 +0,34 Nb1c3
2/8 00:00 140 0 +0,20 Nb1c3 Nb8c6
3/8 00:00 353 0 +0,30 Nb1c3 Nb8c6 Ng1f3
4/13 00:00 636 0 +0,20 Nb1c3 Ng8f6 Ng1f3 Nb8c6
4/19 00:00 2.695 0 +0,32 d2d4 e5xd4 Ng1f3 Ng8f6 Qd1xd4 Nb8c6
5/19 00:00 3.967 0 +0,10 d2d4 e5xd4 Ng1f3 Ng8f6 e4e5 Nf6e4 Qd1xd4
5/19 00:00 4.853 0 +0,30 Nb1c3 Ng8f6 Ng1f3 Nb8c6 d2d4
6/19 00:00 7.096 0 +0,24 Nb1c3 Nb8c6 Ng1f3 Bf8c5 Nc3d5 Ng8f6
7/19 00:00 12.074 0 +0,32 Nb1c3 Ng8f6 Ng1f3 Bf8b4 Nf3xe5 d7d6 Ne5f3 Bb4xc3 d2xc3 Nf6xe4
8/19 00:00 15.287 0 +0,27 Nb1c3 Ng8f6 Ng1f3 Bf8b4 Nf3xe5 00 Bf1c4 Nb8c6 Ne5xc6 d7xc6
9/20 00:00 20.026 0 +0,24 Nb1c3 Ng8f6 Ng1f3 Bf8b4 Nf3xe5 d7d6 Ne5f3 00 d2d3 Nb8c6
10/21 00:00 25.577 0 +0,24 Nb1c3 Ng8f6 Ng1f3 Bf8b4 Nf3xe5 00 d2d3 d7d6 Ne5f3 Nb8c6
11/24 00:00 43.189 0 +0,24 Nb1c3 Ng8f6 Ng1f3 Bf8b4 Nf3xe5 00 d2d3 d7d6 Ne5f3 Nb8c6 Bc1e3
11/24 00:00 86.360 0 +0,29 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 c7c6 d2d3 Ne4f6 Nb1c3
12/24 00:00 98.732 0 +0,24 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1f4 Bc8f5
12/25 00:00 119.869 0 +0,26 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1f4 Bc8f5 000
13/25 00:00 125.026 0 +0,25 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1f4 Qe7xe2+ Bf1xe2 Bc8f5 000 Bf8e7
13/29 00:00 163.325 0 +0,26 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1f4 Bc8f5 000 Qe7xe2 Bf1xe2
14/27 00:00 280.778 0 +0,29 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Bc8f5 Nf3d4 Bf5g4 Qe2xe7+ Bf8xe7 h2h3 Bg4d7
15/27 00:00 414.827 0 +0,29 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 c7c6 d2d3 Ne4f6 Nb1c3 g7g5 Qe2xe7+ Bf8xe7 Nf4e2 h7h6 Bc1e3
16/27 00:00 477.580 0 +0,23 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 c7c6 d2d3 Ne4f6 Nb1c3 d6d5 Qe2xe7+ Bf8xe7 Bf1e2 g7g5 Nf4h5 Nf6xh5 Be2xh5
16/29 00:00 702.479 0 +0,27 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1f4 Bf8e7 Nf3d4 Bc8d7 00 Nc6xd4 c3xd4
17/35 00:02 2.173.083 7.488.604 +0,23 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 c7c6 d2d3 Ne4f6 Nb1c3 d6d5 Qe2xe7+ Bf8xe7 Bf1e2 d5d4 Nc3e4 Nf6xe4 d3xe4
17/32 00:02 2.279.391 7.488.604 +0,25 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1e3 Bf8e7 Nf3d4 Nc6e5 Nd4b5 Ke8d8 Nb5xa7
17/35 00:02 2.882.352 7.310.980 +0,29 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 Ne4f6 Nb1c3 Nb8c6 Nc3d5 Qe7xe2+ Bf1xe2 Ke8d8 Nd5xf6 Nc6d4 Be2d3 g7xf6 Nf4d5 Rh8g8
18/38 00:07 6.802.019 7.066.779 +0,24 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1f4 Bf8e7 Nf3d4 Bc8d7 Ke1d2 Nc6e5
18/40 00:17 15.673.444 6.952.083 +0,31 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 Ne4f6 Nb1c3 Nb8c6 Nc3d5 Qe7xe2+ Bf1xe2 Ke8d8 Nd5xf6 Nc6d4 Be2d3 g7xf6 Nf4e2 Nd4c6 Bd3e4 Bc8e6
19/48 00:41 37.837.204 6.933.179 +0,22 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1e3 Bf8e7 000 Bc8e6 Kc1b1 00 Nf3g5 Be6f5 Rd1d5
19/48 00:41 36.256.767 6.933.179 +0,23 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 Nb8a6 Nf4d5 Qe7e5 d2d3 Qe5xd5 d3xe4 Qd5e5 f2f4 Qe5a5+ Bc1d2 Na6b4 Qe2b5+ Qa5xb5 Bf1xb5+ Nb4c6
19/45 00:41 38.153.057 6.933.179 +0,24 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5d3 Nf6xe4 Qd1e2 Qd8e7 Nd3f4 Ne4f6 Nb1c3 Qe7xe2+ Bf1xe2 Nb8c6 Nc3d5 Ke8d8 Nd5xf6 Nc6d4 Be2d3 g7xf6 00 Bf8h6 Nf4e2 Nd4c6 Bd3e4
20/52 02:31 130.522.258 6.951.227 +0,19 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Nb8c6 Bc1f4 Bc8f5 000 000 Kc1b1 d6d5 Nf3d4 Nc6xd4 c3xd4 Qe7e4 Qe2d2
20/48 02:31 136.857.844 6.951.227 +0,22 Ng1f3 Ng8f6 Nf3xe5 d7d6 Ne5f3 Nf6xe4 Qd1e2 Qd8e7 Nb1c3 Ne4xc3 d2xc3 Qe7xe2+ Bf1xe2 Nb8c6 Bc1e3 Bf8e7 000 00 Kc1b1 Bc8g4 h2h3 Bg4f5 Be2c4

[rge]

Ran 4.43 again, this time with all other software off, including real temp, just turned it on at end for screen shot.