AMD Zambezi news, info, fans !

[xVeinx]

Savantu is one of the resident AMD haters (and was banned from the news section for trolling the AMD threads I believe), so I do recommend taking anything he says regarding AMD with a fairly large amount of salt

I don't see anything he's saying that isn't correct at some level. A bit pessimistic perhaps, but true nonetheless .

[sviola]

I see many say they would be happy with performance between 2500K and 2600K. Well if you look at 1100T's performance in the test suite that will be used for BD( think most MT benchmarks) you will see that 1100T is already in between 2500K and 2600K,so I don't understand this point at all. 8150 will be in the same perf. slot as 1100T is today.,maybe just slightly better overall due to much higher clock speed. Use applications and games with lower thread count and you will see how 8150's weak cores stumble and fail to outperform even Thuban while having extra clock speed advantage. This is fail ,no excuses.


Is this your wish or you have any source to back this up?

MS TechNet is the source: http://technet.microsoft.com/en-us/s...letin/ms11-oct

Edit: forgot to quote the message...

[tbone8ty]

I score 7.03 in cinebench 11.5 with my 1090t @ 4.0 ghz

There's no reason why an 8 core bulldozer shouldn't score at least a 7

This is all getting to be ridiculous

If leaked benchs are correct im all set I'll pass.

Here's to holding out hope it's just a bios/windows bug

[Isbre]

Tomorrow apparently is patch Tuesday, so maybe Microsoft will release an update to W7, which changes windows scheduler? So we will maybe see performance like in the slide AMD showed at computex when they said BD running on windows 8 developer preview, there would be I think 10% FPS increase in L4D2 because of that? And 5 % in other games.

[PetNorth]

It seems Bulldozer won't be remotely what folks thought, but why do you compare Bulldy and Thuban clock to clock? They are totally different arquitecture......

[tbone8ty]

Because if bulldozer arch is this bad Amd was better off making phenom II X6 at 32nm with higher clock

Lol but I digress it's only one benchmark

[Hondacity]

It seems Bulldozer won't be remotely what folks thought, but why do you compare Bulldy and Thuban clock to clock? They are totally different arquitecture......

because amd said the new generation is better than the old generation(bluntly speaking)

[savantu]

Surely you don't think it is as scummy as paying off Dell etc to not use AMD chips, slow other chips in their compilers etc. Or for that matter, willy nilly deciding which ISA is supported or not (remember SSE5, FMA4). If you talk about X86 compatibility, then Intel has a lot more to answer than other two companies involved.

Pointing someone else's failures doesn't eliminate yours. Intel is guilty of all that. How does that nulify the fact that AMD PR is currently in damage control mode, trying to spin the whole matter in the most positive light ? The whole "wait for trinity/piledriver" movement is already under way and BD hasn't been released yet !

Also, wait till NDA lapses before you decry the chip. As it is, we all know how Nvidia screwed DX10. We also have recently learned that windows is more optimized for HT than actual cores, which is to be fixed. So either you could be sane, and rational, or just jump around tearing up your hair.

I don't need an NDA lifted to make an opinion on the chip. I have the same opinion since 18 months and I will be proven right. Talk about consistency, a rare treat these days. I've been banned, insulted, laughed at, whatever. Reality however shows I interpreted the signals correctly and most others ( except a handful ) were day dreaming. No problem, reality will crash their dreams. BD will not touch SB, nice idea, lousy execution and too many tradeoffs.

Savantu is one of the resident AMD haters (and was banned from the news section for trolling the AMD threads I believe), so I do recommend taking anything he says regarding AMD with a fairly large amount of salt

Yes, the "hater" proven right, I must be damn good if I hit it correctly in "anger mode.". Just FYI, it's never about hate or preferences, whatever. Only idiots love or hate an abstract entity, a corporation. It's only the debate on the technical grounds which thrills me. My only pleasure is in debating, be it CPUs or corn crops, with other smart people. That's the whole purpose of a forum, no ? Exchanging, sometimes heated, ideas and opinions.

[PetNorth]

because amd said the new generation is better than the old generation(bluntly speaking)

That doesn't make any sense sorry.

Better or worse =/ clock to clock

[Hondacity]

not sure if you guys have seen the first post but looking at the first post from flanker..

these czech guys flanker and obr are ...idunno

PC Mark:
x6 1100T 6430b
i7 980x 7150b
i7 2600k 7700b
AMD Zambezi 7950b

Cinebench R11.5
x6 1100T 5.9b
i7 980x 8.95b
i7 2600k 6.7b
AMD Zambezi 11b !!!

3D06 CPU
x6 1100t 5990b
i7 980x 7388b
i7 2600k 6600b
AMD Zambezi 8800

[MaddMutt]

anyione up to date on this ? My win7 and 8 were fully patched, no idea if this is included yes or no

"Actually, we already have such an issue known for Bulldozer, and NO bench-marked system has the patch installed!

The shared L1 cache is causing cross invalidations across threads so that the prefetch data is incorrect in too many cases and data must be fetched again. The fix is a "simple" memory alignment and (possible)tagging system in the kernel of Windows/Linux.


I reviewed the code for the Linux patch and was astonished by just how little I know of the Linux kernel... lol! In any event, it could easily cost 10% in terms of single threaded performance, possibly more than double that in multi-threaded loads on the same module due to the increased contention and randomness of accesses.

Not sure if ordained reviewers have been given access to the MS patch, but I'd imagine (and hope) so! Last I saw, the Linux kernel patch was still being worked on by AMD (publicly) and Linus was showing some distaste for the method used to address the issue""


And Tiborr, sorry but changing all sort of options in the bios is not done by most users, neither is setting affinity or co.. Either it works or it doesn't. Not everybody has Chews knowledge on the matter. Disabling PowerManageralready boosted multithreaded performance on my testbed...

Thank You for this Knowledge

This make's some other info very useful

[tifosi]

Pointing someone else's failures doesn't eliminate yours. Intel is guilty of all that. How does that nulify the fact that AMD PR is currently in damage control mode, trying to spin the whole matter in the most positive light ? The whole "wait for trinity/piledriver" movement is already under way and BD hasn't been released yet !

I don't need an NDA lifted to make an opinion on the chip. I have the same opinion since 18 months and I will be proven right. Talk about consistency, a rare treat these days. I've been banned, insulted, laughed at, whatever. Reality however shows I interpreted the signals correctly and most others ( except a handful ) were day dreaming. No problem, reality will crash their dreams. BD will not touch SB, nice idea, lousy execution and too many tradeoffs.

Erm, pointing at Intel does matter as they're a monopoly by and large and use their weight to change the rules of the game all the time. What i listed was Intel's attempts to slow not just the competition down, but innovation in X86 computing. Haswell includes some stuff AMD is doing today and were ready with 2-3 years ago, and you could see blue team screaming like little girls about it. However, no one acknowledges who brought those technologies to fruition. Hey, if you think Intel's ok doing so, then AMD can be allowed to make a bad chip or two sometimes.

[liberato87]

not sure if you guys have seen the first post but looking at the first post from flanker..

these czech guys flanker and obr are ...idunno

which post are you referring too?

[gosh]

which post are you referring too?

http://www.svethardware.cz/forum/sho...4&postcount=27

[wez]

http://www.svethardware.cz/forum/sho...4&postcount=27

Keep in mind that this is from January

[LightSpeed]

I really wanna play with BD and I hope to god that it performs atleast 10% faster overall and im ready to grab a 8120 then.

SB is certainly fast but boring as .

[EniGmA1987]

SB is certainly fast but boring as .

Exactly why I sold my sandy bridge computer. Bulldozer will still be powerful enough for anything out there today even if it doesnt win in every benchmark. I just want to have fun overclocking a computer again.

[BeepBeep2]

not sure if you guys have seen the first post but looking at the first post from flanker..

these czech guys flanker and obr are ...idunno

OBR and flanker are not related, OBR has been preaching about how bad BD is for quite some time now.

It's a funny thing, because his numbers posted months ago now match what we saw from Lab501 and other leaks. These are numbers from OBR's "crippled CPUs"...
And this thread started out as a bunch of crude guesses of numbers from flanker pulled out of thin air.

[spurious]

Does the NDA expire at midnight US eastern time, or some other time zone?

[Hondacity]

Does the NDA expire at midnight US eastern time, or some other time zone?

the nda is under nda lol

[Manicdan]

Does the NDA expire at midnight US eastern time, or some other time zone?

i somehow think that detail is covered in the NDA... and thus we will know after such event happens

[Manicdan]

double posted
and beating by a honda

[muziqaz]

Its 10th today, so still two days or so

[FlanK3r]

lol, what about me guys? This slide is here at first page of this big thread. It was calculation from this slide, because we know numbers for i7 and x6 1100T. Informal remember it...

[LightSpeed]

Exactly why I sold my sandy bridge computer. Bulldozer will still be powerful enough for anything out there today even if it doesnt win in every benchmark. I just want to have fun overclocking a computer again.

This. I was using a X3 > X4 and overclocking it to 3.8ghz stable was extremely fun. I got stuck at 3.3, then 3.4 then 3.5 with seemingly no way past that then voila, one more tweak and i was past 3.6 and so on. Then started with the nb and overall it was the most fun ive had in years when it came to overclocking. SB? Changed multi, upped volts and within 10 minutes i was at 4.6 with the max being 4.7. Fast, effecient but as i said, boring as .

I need to bring a fairy dust good for 10% more performance and blow it on the 8120 I buy. I just hope AMD supplies the dust by day after tomorrow cuz I cant seem to find it anywhere.