Kyosen, in fucntion of this post, when you have the time, no pressure ofcourse. could you test that as well the max fsb? i know the board may be limiting but just to get an glimpse :DQuote:
Originally Posted by BeardyMan
thx for your efforts :up:
Printable View
Kyosen, in fucntion of this post, when you have the time, no pressure ofcourse. could you test that as well the max fsb? i know the board may be limiting but just to get an glimpse :DQuote:
Originally Posted by BeardyMan
thx for your efforts :up:
I also have interest for max FSB.Quote:
Originally Posted by BeardyMan
This board, KFSN4-DRE, has no HyperTransport multiplier setting in BIOS,
so I should be restricted...
...but now I can change multiplier for each core with CrystalCPUID:)
http://222.151.153.254/c-board/file/...iplier-mod.png
Many thanks to Franck, I've inspired and received hints by chatting with you:up:
I think that I should be able to change NB's multiplier, Vcore, Vnb,
within maximum limit which each CPU product has...not yet tested, though.
I see, so with this info you would be able to push more out of it?Quote:
Originally Posted by kyosen
K10's IMC will have the clockspeed equal to cpu clockspeed when installed on old AM2 or 1207 mobo;Quote:
Originally Posted by v0dka
and it will have its own clockspeed on AM2+ or 1207+ mobo
as you can see kyosen's mobo adjusted NB speed to 1800 MHz - IMC is slower than cpu. Right decision - server cpu has slow memory, so it doesn't need high-speed IMC (667*2=1333 MHz will be enough)
But in Phenom's case 1800MHz won't be enough - dual-channel ddr2 1066 needs at least 1066*2=2133 MHz, so I expect Phenom's IMC will run >2200 MHz and it's clockspeed will be adjustable - manually or auto
ht link speed multiplier.. require ldt_stop.. imposible to change it in windows.. if this chipset doesnt support ht3. it will hold you back..
Are you sure about that? I wasn't aware that any cores had a dedicated part of the L3 -- I thought it was completely shared.Quote:
Originally Posted by MR_SmartAss
The way I read the Real World Tech article, the L3 latency should vary from core to core if they're all running at different speeds. The latency for each core seems to be dependent on how that core is clocked relative to the L3, but I'm curious about how this shows up in practice.
Now I can control NorthBridge(Integrated Memory Controller) and HT Link multiplier
with good old WPCREDIT, in Windows:)
Yeah, I'm surprised, tictac...but Everest and CPU-Z tell so at least.
NB @ 200x9, K10(B1) 2350 original
http://222.151.153.254/c-board/file/...tiplier_x9.png
NB @ 200x8, modified on 2350
http://222.151.153.254/c-board/file/...tiplier_x8.png
And HT Link.
Originally, Everest and CPU-Z are showing different HT Link clock:
http://222.151.153.254/c-board/file/...T_x1_x5_x1.png
F0xC8 modified...now CPU-Z also shows 1000MHz = 200x5!:
http://222.151.153.254/c-board/file/...T_x1_x5_x5.png
F0x88 modified too...nothing occures, but it still works, at least:
http://222.151.153.254/c-board/file/...T_x5_x5_x5.png
And all 3 HT Link can be set at 600MHz = x3 multiplier:
http://222.151.153.254/c-board/file/...T_x3_x3_x3.png
Now I should try max FSB test...but very sleeeeepy.....good night;)
how much for those Opterons?
ok.. i eat my word.. lol.. its only 12midnight here.. come on.. share the might..
do you have a vid for that?:DQuote:
Originally Posted by tictac
just kidding :p
thx for sharing kyosen, looking forward to teh results ;)
kyosen : good work there :) , and thx for sharing to bad you need to sleep :(
Those 2 2350's are my personal pair.....you should have asked a couple weeks ago as I had a *few* which needed homes :D.Quote:
Originally Posted by mad_skills
no matter, i'm waiting for X4, i was just wondering how much were they going for when you bought them..
Look forward to max FSB test. With the possible voltage mods as well should be interesting. Hats off to Kyosen ( with Mr CPU-Z's assistance :) looks like some interesting results to come.
Shame the initial scores are 10% down on Kentsfield, I was hoping it would be 10% up to take up the slack of not being able to overclock as well. Hmm, maybe on final Spider release it will be about level.
Regards
Andy
Max fsb will be interesting indeed.
Here Phenom reached as much 3.3Ghz already. Although that is on a different platform and seems to be done by multiplier overclocking, I wouldnt be surpised if these Barcelona's had a lot more in them.
Well spotted v0dka. :up:Quote:
Originally Posted by v0dka
3 builds from now, a few weeks and I hope I'll be able to test one an find out for myself. Lets hope 690G chipset works with these ... at least initially. :)
"Pictures of AMD Phenom 9600 processor
Pictures of AMD Phenom 9600 processor and CPU-Z screen shot. The name of the CPU is Phenom X4 with codename Agena. It is rated 65nm at a core voltage of 0.976v. The processor is labelled Phenom 9600 clocked at 2.3GHz. The CPU is installed on a NF560 chipset mainboard, that is the reason you see the HT is set to 1GHz. It has 512K L2 cache and 2M L3 cache."
Vcore only 0.976v o_0 Aint that looking good for the Phenom ?
http://my.ocworkbench.com/bbs/showthread.php?%20threadid=68538
Yep,very low Vcore for 2.3Ghz quad core part :).Where is savantu with his 1.52V claims for 2.5Ghz K10s :D?
This could mean that for 3Ghz,one would have some serious headroom!
BTW ,this vcore is lower than the ones intel specced for 45nm C2Qs at the same freq.
link is busted Det0x :(
Fixed it now :)Quote:
Originally Posted by SparkyJJO
thats pretty exciting! My x2 4000+ needs 1.025v for 2.1ghz...they've double the cores and upped the frequency versus the 4000 by 200mhz, and use .5v less. Performance/watt!
This should come in handy for tweaking, if you can find it:
http://forums.vr-zone.com/showthread.php?t=199812
That is very nice headroom for a 2.3G part. 3G and at least one core at 3.3G. I do hope that's not going to be the highest end Phenom that will be comming out this year though. Was hoping for at least 2.6G myself.
At 0.976V Vcore, I seriously doubt it's the top.Quote:
Originally Posted by PhilDoc
Tweaking the all the different clocks could be quite a playground!
I'm thinking the voltage of 0.976V is probably 1.0V rated, sometimes CPU-Z reads a bit off (reads mine slightly low on my opteron 165 as well).
Quote:
Originally Posted by j3pflynn
The tweaking part would make it more fun for my money. The only reason I worry is that the chip seems to be sent out for review, although, I could be mistaken. Anyway, I'm sure that they will debuy higher.
That's a B2 step and very low voltage for a quad at 65nm. That's not ES either. Looks like it'll be multi's of 11.5x 12x 12.5x 13x and maybe 13.5x and 14x too. IF the CPU can get to 300HT, what a blower that'd be with these multi's. :D
Top part is higher than 2.5GHz. IIRC it's supposed to be a 2.8GHz for December but in October that depended on future production so they will probably have highest working part for release but very limited quantities (like QX9650).
with all that power coming from the B2s, what can we expect from the opterons and the fasn8 plataform?
Btw, does anybody has any info if fasn8 will use HT1.0 or HT3.0?
so many contradicting infos lately eh... Phenom slower than K8 in Crysis,
1.5V for 2.5ghz, 1.2V for GP9600 and now 1.0V for teh same GP9600... arghhh :S
though this '1V for 2.3Quads' reminds me of that saying: 'At 3ghz, the Phenoms
are stone cold killers, much faster than anything AMD and Intel has to offer today'
of Rahul Stood :D
A real pleasure to read your thread kyosen, as usual you have some rather interesting info we can trust 100%. Also thanks to stephen for sending these chips on vacation :up:
I had been a bit busy, sorry.
Max FSB is around 245 for 2350(B1) on my KFSN4-DRE so far,
still Vcore&Vnb is 1.20~1.25V area, though...
...I intend to do voltage mods on this server board:)
BTW, Franck/cpuz is working hard for K10 support:up:
Now CPU-Z can show NB clock, each core clock/multiplier, memory acccess ganged/unganged...
http://222.151.153.254/c-board/file/...1_unganged.png
...and memory clock calc method is changed for K10...ratio against FSB, same as Everest.
It means that I had compared K10(BA) at 333MHz memory clock and K8 at 300MHz memory clock...
...comparison for K10(B1) and K8 was no problem, both were 333MHz memory in stock setting.
Then I tried again SuperPI4M for K10(B1) and K10(BA) at same core/NB multiplier.
I had expected ~1% = 1~2sec difference, but there isn't such a difference.
So, I was misunderstanding...now I should say B1 and BA is almost same performance,
at least for SuperPI.
Yeah s7e9h3n was/is right:)
So, the conclusion is we can fit barcelona's performance less than Quad Intels and greater than dual opterons/Xeons ??
OR
Yet to see the B2 stepping ?
some thing else ?
PS: No offense ...trying for a honest opinion.
seems that B1 and BA are "almost" the same so.
My questioning would be: in which cases does it perform better, and which does it performs worst?
And: how much better or worst? ;)
is B2 coming out?
Great work, my friend! Ok these questions are for either you or Franck ;) ...Quote:
Originally Posted by kyosen
One thing I'm a bit confused about - I see the CpuZ shots in this picture identify logical cores as an individual processors i.e., Processor #1 and Processor #2. Now what happens when you insert a second PHYSICAL cpu? Will CpuZ assign each core #5-8? Or will there be some way which it differentiates between physical cpu #1 and physical cpu #2?
Will CpuZ be capable of reading the Vcore on Barcelona? To this point, the only tool that I believe accurately depicts Vcore is AMD's power monitor. It'd be great if CpuZ could do the same.
Hi S7e9h3n,
I'm confused as well. This should not report this way, but one processor with 4 cores / 4 threads, as on previous screenshots.
Kyosen is modifying MSRs, so this can make cpuz wrongly detect some things, this may explain. He will help me figuring out anyway.
AMD's power monitor reports the Core VID, I'm currently adding that in cpuz as well. But the real VCORE is always better, and in this case cpuz did not detect the sensor chip.
I'm also preparing a small tool to change multiplier on the fly for each core. Kyosen already does it, but this is not very convenient. A slider will make things much easier for him.
OK...thanks for clearing that up. :up:Quote:
Originally Posted by cpuz
BTW, if you need some data/dumps from a system running 2 x Barcelonas, feel free to send me a PM (of course that's dependent on when my system comes back from repair :rolleyes: )
I presume multiplier is upwards locked in new Barcelona's?
Also is it possible to set vcore higher than 1.2V using C'n'Q or this is maximum allowed?
What is maximum vcore for northbridge?
Can you use half multipliers on Barcelona? Phenom 9600 is running 11.5x so it looks like Barcelona should be capable of doing same. If yes will we see any performance drop from using .5x as on K8?? (mainly lower memory bandwidth).
Lots of questions for our Guru's! :)
Yes :(Quote:
Originally Posted by Lightman
Max VID is 1.55V.Quote:
Originally Posted by Lightman
This depends on dual-plane or single-plane configuration, and I still miss the register values to tell. On Kyosen system, the NB VID is 1.2V on the B1 and 1.15V on the BA.Quote:
Originally Posted by Lightman
Yes.Quote:
Originally Posted by Lightman
I don't know. On K10, memory clock is computed from bus speed, and it therefore not dependant from CPU speed anymore, so this is tempting to say "no", but only a real test can tell.Quote:
Originally Posted by Lightman
Is it possible to figure out the value by looking at which register Amd's Power Now tool is accessing? Also, I think the B1 vs. B2 NB VID's are different because Kyosen's cpu's are HE's :)Quote:
Originally Posted by cpuz
Thank you for your answers Franck! :up:Quote:
Originally Posted by cpuz
With 0.5x multipliers I'm also tempted to say it will not make any difference, but ....
Regarding vcore is it then possible to set 1.55V from Windows?? You know that X2 are hard locked between 1.30 and 1.45V depending on model and version....
BTW have you all guys noted AMD's official pricing for Phenoms??
It is what I was waiting for :) .
Pricing is very competitive ;)Quote:
Originally Posted by Lightman
source?Quote:
Originally Posted by Lightman
:banana::banana::banana::banana: dood, you know we're lazy! where's the linkage?Quote:
Originally Posted by informal
Here's your "linkage":Quote:
Originally Posted by zornundo
http://www.xtremesystems.org/forums/...93&postcount=1
I just got Tyan S2915-E. Seem oc a little bit higher. Now i can run 225 mhz htt.:D. But can not reach higher than 230mhz. Seem mainboard need a lot of tweak. Barcelona 2.3ghz is coming, i saw preorder on web now.:D
Very good news.
And my client confirm One barce 1.9ghz faster than Two opteron 280 in web & streaming servive in their datacenter.
And desktop software do not gain much and using full barcelona power. Becos i saw no gain when using 3dmax9 render(single chanel vs numa enable).
I miss the value of the register, not the register.Quote:
Originally Posted by s7e9h3n
This register is not "standard" and the generic dumps do not show it. It needs a special access. The next test will answer to that question (max NB VID)
as far as i can understand according to kyosen's SS:
[IMC clock]:[phys. memory clock] == 27:5 when IMC clock == 1800 and ddr2 667 installed
and
[IMC clock]:[phys. memory clock] == 24:5 when IMC clock == 1600 and ddr2 667 installed, etc.
and respectively
[L3 cache clock]:[phys. memory clock] == 27:5 when IMC clock == 1800 and ddr2 667 installed
and
[L3 cache clock]:[phys. memory clock] == 24:5 when IMC clock == 1600 and ddr2 667 installed, etc.
right?
SuperPI4M time with different NB clock
2350(B1) @1.8G=200x9, NB@1.8G=200x9: 3m47.453s
http://www.oohashi.jp/c-board/file/S..._NB1.8_UnG.png
2350(B1) @1.8G=200x9, NB@1.6G=200x8: 3m47.359s
http://www.oohashi.jp/c-board/file/S..._NB1.6_UnG.png
Almost same time...difference is within fluctuation, I think.
NB@1.6G may be enough for single thread program with dual DDR2-667 access!?
or I suspect my NB multiplier changing method is incomplete:(
I have no detail info about L3 clock and its behaviour...
3DMark06 score with different HT Link clock
2350(B1)@2.0G, GeForce8800GT@default clock, HT Link@1000MHz=200x5: 9418
http://www.oohashi.jp/c-board/file/3..._HTx5_88GT.png
2350(B1)@2.4G, GeFoce8800GT@default clock, HT Link@962.8MHz=240.7x4: 10926
http://www.oohashi.jp/c-board/file/3..._HTx4_88GT.png
We can observe slight improvement compared to previous results, 9353 and 10796:
http://www.oohashi.jp/c-board/file/3....0G-200x10.png
http://www.oohashi.jp/c-board/file/3....4G-240x10.png
K10 support in MchbarEdit and MemSet
FELIX provided me beta programs, both are working fine:up:
Now we can change NorthBridge & HT Link multiplier with MchbarEdit,
and memory timing with MemSet, on Windows...both 32bit and 64bit:clap:
(Good old WPCREDIT has no support for Win x64)
Screenshots are located on my BBS:
http://www.oohashi.jp/c-board/c-boar...ne;no=5239;id=
CineBench10 32bit/64bit
32bit, 2346(BA)@1.8G, NB@1.6G, DDR2-667: 1483, 5731, 3.86 x
http://www.oohashi.jp/c-board/file/C..._NB1.6_333.png
32bit, 2346(BA)@1.8G, NB@1.6G, DDR2-800: 1487, 5730, 3.85 x
http://www.oohashi.jp/c-board/file/C..._NB1.6_400.png
32bit, 2350(B1)@1.8G, NB@1.6G, DDR2-667: 1484, 5743, 3.87 x
http://www.oohashi.jp/c-board/file/C..._NB1.6_333.png
32bit, 2350(B1)@1.8G, NB@1.6G, DDR2-800: 1493, 5736, 3.84 x
http://www.oohashi.jp/c-board/file/C..._NB1.6_400.png
32bit, 2350(B1)@2.4G, DDR2-640: 1973, 7654, 3.88 x
http://www.oohashi.jp/c-board/file/C..._NB1.8_320.png
32bit, 2350(B1)@2.4G, DDR2-800: 1977, 7672, 3.88 x
http://www.oohashi.jp/c-board/file/C..._NB1.8_400.png
64bit, 2346(BA)@1.8G, NB@1.6G, DDR2-667: 1793, 7315, 4.08 x
http://www.oohashi.jp/c-board/file/C..._NB1.6_333.png
64bit, 2350(B1)@1.8G, NB@1.6G, DDR2-667: 1789, 7327, 4.09 x
http://www.oohashi.jp/c-board/file/C..._NB1.6_333.png
64bit, 2350(B1)@2.4G, DDR2-800: 2402, 9642, 4.01 x
http://www.oohashi.jp/c-board/file/C....4_x10_400.png
64bit, 2350(B1)@2.3G=230x10, DDR2-768: 2279, 9191, 4.03 x
http://www.oohashi.jp/c-board/file/C....3_x10_384.png
64bit, 2350(B1)@2.3G=242x9.5,DDR2-808: 2306, 9239, 4.01 x
http://www.oohashi.jp/c-board/file/C...3_x9.5_404.png
Summary from above:
* B1 and BA show almost same performance.
* DDR2-667 or 800 makes almost no difference.
* 32bit -> 64bit score gain is about 20%.
* K10 has no performance degradetion at half(*.5) multiplier.
BTW, Franck provides me CPU-Z latest beta:up:
Now we can see K10 VID on CPU-Z(for example, in the 64bit, 2346(BA)@1.8G case above),
and also can choose a clock showing core#:clap:
I intend to test a rendering software, Shade...familiar in Japan,
and then, I'll try voltage modification:)
thanks for all the work kyosen, franck, felix, and others ;)
Why do you think that there is a 20% improvement from 32bit to 64bit?Quote:
Originally Posted by kyosen
actually depending on the taskQuote:
Originally Posted by Eternalightwith
20 per cent is an average mean :cool:
kyosen, and where's 64bit 3dmark?
correct, this (NB FID) can not be changed on the fly (the register can be adjusted but NB clock doesnt actually change). a BIOS item will allow NB multiplier (NB FID) adjustment.Quote:
Almost same time...difference is within fluctuation, I think.
NB@1.6G may be enough for single thread program with dual DDR2-667 access!?
or I suspect my NB multiplier changing method is incomplete
are you sure? then how CnQ will work with IMC?Quote:
Originally Posted by macci
CnQ lowers both CPU and IMC multi in idle, on the fly :confused:
Thanks for your comment:)Quote:
Originally Posted by macci
OK, I read the document again........yeah, it looks that I need 2 more steps, i.e.
1. write register NbFidEn=1
2. then, reboot.
I'll try it tonight!
And MAS, well, I'll also try 3DMark on WinXP x64:)
Thanks to all guys, and kyosen. :)
Kysoen: Are you sure there is almost no difference with different DDR2 clocks in CB10? It should be around 100 CB between 667 and 800.
Sure...yeah it's strange, I feel too.Quote:
Originally Posted by KTE
I got different time for SuperPI4M with different memory clock, as usually expected.
Hope you don't mind me quoting one of your tests Kyson, but I just happend to spot an almost identical 3dmark run on OCAU with a Q6600 @ 2.4:
8800GT+ Q6600@ 2.4 : http://imagestore.ugbox.net/aview/3D...3db6c787b03544
11174 3dmarks
SM2:4651
SM3:4764
CPU: 3473
8800GT+ 2350@ 2.4: http://www.oohashi.jp/c-board/file/3..._HTx4_88GT.png
10926 3dmarks
SM2:4439
SM3:4672
CPU: 3553
that's a nice one by 2350 :up:
looks like Opty is going to be a good one..:)
You've mixed up the global score ; the Intel system scores 11174.Quote:
Originally Posted by mAJORD
That's because(the higher intel score) the intel system was benched on Vista while AMD was on win 2003.The driver was obviously better on Vista(2003 is kinda old) and don't forget the HT speed the Barcelona was running with, along with regged mem. with cas 5.Quote:
Originally Posted by savantu
The important thing here is that K10 scored better in CPU score dept.
oops, Cheers Savantu. :)
I had all the intel SM scores in the wrong spot before first ninja edit too :/ I'm a bit tired (exuses)
Quote:
Originally Posted by mAJORD
This doesn't count because user with Q6600 uses Windows VISTA.........So.......ADD another 1000marks or so on it for Windows XP and you're ready.... ;)
Bench windows 2003 please and see teh difference, windows vista ai'nt that shabby.Quote:
Originally Posted by hipro5
IF so, so HOW can ANYONE compare?..... :p:Quote:
Originally Posted by BeardyMan
Well , what's obviously for you isn't for others.Quote:
Originally Posted by informal
The 2003 driver is basically the XP one.I'm pretty sure the XP driver is better than the new Vista ones.They had 6 years to perfect it after all.
Secondly , HT speed won't improve scores.
Thirdly , we don't know the RAM settings for the Intel system.Kyosen said : * DDR2-667 or 800 makes almost no difference.
I'm pretty sure that will also happen on desktops.The large latency of the L3 pretty much negates the few ns won or lost by using 667/800/1066RAM plus you have better prefetchers , in other words K10 will be less sensitive to memory latency than K8.
True.Quote:
The important thing here is that K10 scored better in CPU score dept.
Did it use 1 or 2 CPUs ? It is a DP board anyway.
More importantly , this post pretty much makes all the above useless :
http://www.xtremesystems.org/forums/...&postcount=181
BA is B1 for mass production , the differences are none => BA=B1 in performance.Quote:
CPU clock: 3DMark Score / CPU Score
K10(BA) 2.0G=222x9: 9482 / 2999
K10(B1) 2.0G=200x10: 9353 / 2979
K10(B1) 2.2G=220x10: 10148 / 3267
K10(B1) 2.4G=240x10: 10796 / 3560
K10(B1, 3core) 2.4G=240x10: 10420 / 2830
K10(B1, 2core) 2.4G=240x10; 9423 / 1888
Kentsfield 2.4G=267x9: 11901 / 3845
Kentsfield 3.0G=334x9: 12965 / 4792
Quick conclusion from above:
*K10(BA) looks a bit faster than K10(B1), though multiplier aren't same.
*K10 score seems about 10% lower than Kentsfield at same clock.
So, K10 needs more clock!!!
I have own Kentsfield 3DMark06 results...
...haven't you seen my old post in this thread!?;)
http://www.xtremesystems.org/forums/...&postcount=181
I pick up them and list along with revised K10 results:
CPU clock: 3DMark Score / CPU Score
K10(B1) 2.4G=240x10: 10796 / 3560 -> 10926 / 3553 (with HT multiplier x4)
http://www.oohashi.jp/c-board/file/3....4G-240x10.png
http://www.oohashi.jp/c-board/file/3..._HTx4_88GT.png
Kentsfield 2.4G=267x9: 11901 / 3845
http://www.oohashi.jp/c-board/file/3...2.4G_267x9.png
So, Kentsfield is ~10% faster than K10, for both Total and CPU score...
Edit: Ah, savantu is faster than me:)
Hmm strange,i went by HKEPC numbers here for yorkfield:
http://www.hkepc.com/?id=171&page=3#view
It seems your kentsfield is ~10% faster than Penryn quad used in HKEPC with 3% higher clock(2.4Ghz for your Kentsf. vs 2.33Ghz HKEPC Yorkf.)
You haven't got a RD790 board...:DQuote:
Originally Posted by kyosen
On one side (kentsfield) this is ddr3, on the other (K10) it is ddr2 Is 3Dmark and cpu score dependent of memory?
What VGA HKEPC used?....:)Quote:
Originally Posted by informal
Quote:
Originally Posted by kyosen
Ahhhhh.....kyosen-san, you're cheating........:D :p:
WHY did you run the Barchie with HIGHER fsb than it's default (200MHz)...?... :p:
Now to be equal, run Kendsfield ASLO at higher fsb (for example at 333MHz fsb)...... :D :p:
Memory bandwidth/latency has virtually no impact on the 3DMark06 CPU score:Quote:
Originally Posted by nemrod
A single 3GHz Clovertown scores higher than a QX6800 despite being paired with FB-DIMMs.
http://www.xbitlabs.com/images/cpu/i...8/3dmark-2.png
A few other reviews with dual-cores.
http://www.xbitlabs.com/images/memor...I/3dmark-2.png
http://www.xbitlabs.com/images/memor...z/3dmark-2.png
http://www.xbitlabs.com/images/cpu/a...d/3dmark-2.png
High fsb and high rams frequency helps a bit CPU score in 2k6....;)Quote:
Originally Posted by accord99
You should choose proper word:cool:Quote:
Originally Posted by hipro5
No , but Core based CPUs perform better with DDR3 even if clocked at a miserable 1333@CL8.Quote:
Originally Posted by nemrod
Either way , you compare what is going to be mainstream next year.
It's all about the chipset and RAM used. ( DDR3 > DDR2 )Quote:
Originally Posted by informal
What's the config on Hkepc ?
Ahhh.......Yes, Sorry....... :DQuote:
Originally Posted by kyosen
Hi there,
i have two questions.
1. Where can i get the Bios Version 1004 for my KFSN4-DRE?
2. What is the version of clockgen you use?
I try to overclock my Opteron 2344 and the system freeze when i apply the settings in clockgen.
Greetings from germany
indiana_74
You are welcome:)Quote:
Originally Posted by hipro5
And indiana_74, I've just read your private mail...give a moment;)
Welcome aboard dude :welcome:Quote:
Originally Posted by indiana_74
Welcome indiana_74, we are grateful to have you here (and your chips too ;) )
that would be NB DID (divides the clock by 2). but the actual multiplier (NB FID) can not be changed (on the fly).Quote:
Originally Posted by MAS
George,Quote:
WHY did you run the Barchie with HIGHER fsb than it's default (200MHz)...?...
I know your not an AMD guy (:D) but please note that the HT Multiplier in this benchmark is at X4 mode. So the "FSB" is actually underclocked (960MHz vs 1000MHz which is default for that CPU).
The HT ref. clock (240MHz vs 200MHz) makes absolutely no difference in terms of performance.
HT Link (1GHz by default on this example) is the "FSB" of AMD world and it depends on HT Link multiplier and HT ref. clock. so 240x4 (960MHz) is slower than 200x5 (1GHz) (assuming that the Memory bus and CPU core clock would be at the same frequency).
very helpfull comment macci, thanks for posting
surely you're joking. i know you've messed around with plenty of amd chips during K8's reign, and have seen first hand how the htt (or fsb if you want to call it that) does NOTHING for performance ;)Quote:
Originally Posted by hipro5
edit:
should have read the rest of the thread before posting :lol:
good explaination macci, though i'd like to add something else to that: the 'HT Link' speed, or 'FSB" (i get frustrated by how things get multiple names that sometimes overlap) does very little for performance as well, except in multi-cpu situations when running multithreaded apps that require lots of communication between the cpus :D
Thakn you Kyosen for answering on my question (about 0.5 multipliers)! :up:
One thing which strikes me is Barcelona scaling under Win XP x64. It is more than 4x!:confused: How come?
Is this because of the way memory controller is working (unganged mode) or L3 Victim Cache is helping here? Any reasonable explanation??
Well......I asked a friend (user) to our forums who owns a Q6600 and a 8800GT to bench 2k6 at 2400MHz and VGA 600/900......
There you go..... :)
He used "bios defaults"....Ram "by default" , etc.....As a "regular user" is going to buy a system like this and bench 2k6 without overclocking, playing with the rams, etc.....
Barchie 2400MHz (with higher fsb than default - edit: OK HT lower :D ) : 10926 marks...
Core 2 Duo QUAD Q6600 ALL DEFAULT : 12073 marks...
SO:
Barchie
SM2 = 4439
SM3 = 4672
CPU = 3553
Q6600
SM2 = 5052
SM3 = 5056
CPU = 3857
Now what?.... :)
:D I know........BUT I see "FSBs" all over the place....... :D :rofl:Quote:
Originally Posted by macci
That's about right where it should be. Odd that CPU/MB made so much difference to SM2/SM3 scores though, 1000 points.Quote:
Originally Posted by hipro5
http://www.tweaktown.com/reviews/1210/5
http://www.bootdaily.com/index.php?o...1&limitstart=1
http://vr-zone.com/articles/Nvidia_G...w/5369-13.html
It's not that odd......The other user used Windows Vista......;)Quote:
Originally Posted by KTE
wow.. in 64bit windows barcelona speedup.. 4.0x.. thats 100% efficiency.. would be interesting to see other benchmark running on 64bit windows.. 3dmark ?
Nah , that's impossible.Quote:
Originally Posted by tictac
The only explanation is IMO that something is wrong/suboptimal when running only one thread.Which is weird since one thread should have all the L3 for itself.
Moreover , one thread has both IMCs working for it.
In the multithreaded scenario , data BW from both L3 and memory is shared => welcome to Amdahl's law.
http://en.wikipedia.org/wiki/Amdahl's_law
Hate to say it, but you're right......here's my Cinebench on Server 2k3 x64. Kyosen-san's single threaded score is unusually low....Quote:
Originally Posted by savantu
http://img509.imageshack.us/img509/1...ench64akl3.jpg
*BTW, K10 has only ONE IMC, not two ;)
i guess 3.1ghz phenom b2 could break 12k on cinebench under 64bit mode...
A theoretical prediction used only as an estimate of expected performance which has been disproven in many computing scenarios well known among scientists, especially when you integrate caches and variate sub-components in a given processor, and even more so with multi-core technologies. Microprocessors depend on how well they are utilized by a given application and we do not know how to utilize 100% perfect working power and resources of a CPU in all scenarios since the application coding is not CPU specific, but rather wildly generic. Hence, if you could make the coding very optimized and processor specific, you are bound to see a speedup even at the same frequency. It's more about processor resource efficiency.Quote:
Originally Posted by savantu
Disagree. Amdahl is an estimate, but it's maximum performance estimate. if that thing would go 4 times faster in 4 cores, that would need to mean perfect scalling, coding, and even worst, parallel from the first to last line of code.
I really don't see it happening.
Umh...there are actually 2 64 bit channels which put together act as a 128 one.This improves performance when multiple threads access memory.Quote:
Originally Posted by s7e9h3n
This is something Intel chipsets have done since the I865.
Give 2 examples which respect the conditions imposed by the law.Quote:
Originally Posted by KTE
@s7e9h3n
hi my Quad 6600 B3 :D is faster your barci:rofl:
http://s6.directupload.net/images/071105/fd5j463u.jpg
Quote:
Originally Posted by siyah
Dear friend....
I will not ask about your age...:D
But DO YOU REALLY know what means COMPARE?:cool:
Try again with more equal speeds & FSB and post again....
Don’t fire if you don’t know how to use a gun....(you can kill yourself and our sense of humor):rofl:
Hmmm is it becuse the lower speed for the K10 thats make it to have Multiprocessor speedup 6.79x instead of low 3.46x like the Q6600 have? or do i see big improvement with more cores on K10?
6.79x is with 8 cores.Quote:
Originally Posted by Lastviking