Immediately?Quote:
Originally Posted by Olivon
Printable View
Immediately?Quote:
Originally Posted by Olivon
Yes, in corporate time. You know, few meetings, few responsibility dodges etc.Quote:
Originally Posted by gallag
Not true. Many companies in history have released inferior products to replace good ones. Happens all the time. It shouldn't, but it does, and sometimes on purpose.Quote:
Originally Posted by Alex-Ro
Anyone still remember brisbane ? Half the cache, higher lat, and lower clocks than it's predecessor, all this on a new node...Quote:
Originally Posted by Alex-Ro
And first Pentium 4 as well...
True but Brisbane was alot better at undervoltingQuote:
Originally Posted by Kuroimaho
Wow, come on... Hope that's pure luck, or easily hotfixable.Quote:
Originally Posted by duron
No suprise there... I really want to see how the opterons do... if there is a simillar picture it will be devastating for amd...
Quite shocking to look at those graphs, after all these years of waiting for Amd to be competitive...
lol by THOSE metrics, westmere is 50% more efficient than bulldozer, and sandy bridge is 50% more efficient than westmere
sandy bridge PAYS YOU to buy it over bulldozer. unless electricity is free
Is it posible for the adic who ownes gf to devalue amd by holding back there products with a crappy process only then to buy amd for a bargan price? I can see adic buying AMD at some point in the future then gf launching an amazing revision of there current process. wheres my tin foil hat lol.
This error was common with the original Phenom as well. If it wasn't a hot topic then, I don't see why it should be now other than to let people complain more about BD. Shogun players might have a legitimate beef.Quote:
Originally Posted by zalbard
AMD would lose the x86 license in the process, so no that's not what's happening here.Quote:
Originally Posted by gallag
this has got to be the worst possible way to show efficiency.Quote:
Originally Posted by Olivon
first you dont go with the biggest most power hungry motherboards if you want an efficient system
second you dont simply add up times together. some benchmarks took a few minutes, other took a portion of a second. horrible weighting factor and right there shows they are about as dumb as you can get.
third, you show efficiency per app since there are multiple ways to do stuff, and if you really care about power consumption, you pick the app and method that does it best for you.
and last, you might turn off turbo so that you loose a little bit of performance, but gain a huge power return of power savings.
we all knew that these things are power hungry when pushed hard, this is just trying to use shock value to get more hits rather than a proper test.
btw for a quick reference, i was able to increase efficiency on thuban by over 30% by simply dropping the voltage from 1.3 to ~1.1v at stock clocks, and thats with a chip already built around efficiency.
+1Quote:
Originally Posted by Hornet331
I'm looking to see the opteron 6220 vs 8120 FX.
looking to see how L3 cache impacts the cores/ modules.
I would expect the largest gain in the opterons will come from the much more aggressive clockspeed. The current gen tops out at 2.2ghz for the 12 core magny cours and 2.8ghz for the 6 core lisbon. So considering the 8 core parts top out at 3.5ghz (3ghz no turbo) and the 16 core parts top out at 2.8ghz (2.3ghz no turbo) AMD can probably expect some decent gains in integer computation. The weird thing is the FP seems to be much weaker than expected. I can't remember which review it was (i think it was for the 2700k actually) but the fx 8150 had no problems keeping up in integer computation but failed in float point, which is odd since fp had always been AMD's strong suit since the athlon days.
Though the power draw of BD is certainly disappointing, THG was a little sneaky by disabling the Intel on-chip HD graphics and using an HD6850.
That saved 20-30w right there.
Top 12C MC today is clocked at 2.5Ghz.Quote:
Originally Posted by AliG
Quote:
Originally Posted by AbortRetryFail?
That is probably done for fairness. When you can't match up specs exactly since they are different platforms, you still try to make it as similar as possible. The likely purpose of the test was to compare efficiency in a system for workloads(none gaming). Since Zambezi doesn't have integrated graphics, they have no choice but to use a videocard. Intel could run off of the integrated graphics if need be for the test desired.
What would be an unfair test would to have the amd processor use a dedicated, while the intel used its integrated. Since none of these benchmarks are gaming ones, Intel would have the clear advantage since using integrated would probably save more power than using the 6850.
That contradicts with many other test results like:Quote:
Originally Posted by Olivon
http://www.rage3d.com/reviews/cpu/am.../index.php?p=7
http://techreport.com/articles.x/21865
Quote:
These results couldn't be much more definitive. In every case but one, distributing the threads one per module, and thus avoiding sharing, produces roughly 10-20% higher performance than packing the threads together on two modules. (And that one case, the FDom function in picCOLOR, shows little difference between the three affinity options.) At least for this handful of workloads, the benefits of avoiding resource sharing between two cores on a module are pretty tangible. Even though the packed config enables a higher Turbo Core frequency of 4.2GHz, the shared config is faster.
Those results are big news!Quote:
Originally Posted by duron
Why didn't they tests any games ? Bulldozer needs appropriate software to get it ticking !! Dunno if Windows 8 will cut it out of the box...
This was known for some time. Two threads from same module yield only 160% speed increase from a single thread, because with one thread you don't share resources.
^This
AMD said for a long time that you trade off 20% performance to be able to able to add a second integer core at 35% the size
Maybe even better Win 9, 1-5% is not sufficient, sadly enough... let's hope Globol foundries can produce better products soon so we can cranck these CPUs up to where they belong... Seems they made some errors in the charts, putting some worst scores in bold
Seconded, frankly i'm not impressed with those numbers at all. If anything those increases probably come from the windows 8 kernel itself being better, not better task scheduling. We already know from people physically disabling integer cores that ipc goes up by as much as 20% with proper scheduling
True, but how do we know that W8 RTM or even beta won't do the job even better than this pre-beta Windows Developer Preview?Quote:
Originally Posted by Leeghoofd
After all, it's still pre-beta. .
You could just make a benchmark with both 2500k and the amd and see if they both get a 1-5% increase?Quote:
Originally Posted by AliG
What about the Set Affinity feature in the Task Manager? Task Manager -> Processes -> right click a process -> Set Affinity -> Set cores to run process.
guys, guys cmon its over...no more how about this, or if this , if that....It's ALL OVER..BD flopped end of STORY
lol chill out bro. That's obvious, I'm just wondering about the potential performance gains from windows 8. I have a 2500k myself (very glad i caved over the summer and stopped waiting for BD)
people are now trying to figure out the reasons behind the flop of BDQuote:
Originally Posted by RedBull78
not finding excuses.
May I point you in the direction of the AMD PR team?Quote:
Originally Posted by haylui
-PB
How exactly is AMD PR going to discover the issues with with BD? That is like having patients call Pharmaceutical companies for medical advice.Quote:
Originally Posted by paulbagz
nah brah, its just that some are still under the spell of the BD. Somehow or somewhere is a secret patch that will unlock the real BD..get it?
Fair enough analogy. More reason why I dont respect doctors. They should all be required to try the actual med before prescribing you one, so in this case they'll know if its bitter or sweet....on this end we all know how the BD tastes. Most agree it wouldn't hurt the PR team if they really tested the BD to let us know how it tastes...and they did ..they said it tasted OMFG "They're Gr-r-reat " ......."til the PR is gone"Quote:
Originally Posted by Stukov
Clearly missed who and what I quoted lolQuote:
Originally Posted by Stukov
-PB
Yes, I clearly did, hence the question. As your statement dumbfounded me and thus followed the question with the statement that projected that bewilderment.Quote:
Originally Posted by paulbagz
fail post. they would have said the same thing because they were told to say the same thing.Quote:
Originally Posted by RedBull78
:ROTF: :rolleyes:Quote:
Originally Posted by Stukov
-PB
I already cannot wait for windows 8 to come out just so people can acknowledge what a fetid, overrated turd Windows 7 ALWAYS WAS. Vista improved, more than can be said for 7 which is still with us horrid ergonomics and slow boot warts and all.Quote:
Originally Posted by duron
may be socket 2011 hit retail before FX AM3+ in Thailand !!!
Actually,1 BD core is performing somewhere between Phenom I an Phenom II,with occasional dips below and above these. On average it's like Agena core. BUT,it has a much longer pipeline and God knows what else not working as good as it should inside the core. This is taken from PCgameshardware's website:
http://img845.imageshack.us/img845/3572/95136680.jpg
For gaming results with 480GTX click here. Not bad at all for 8150 @ stock. It is definitely better gaming chip than 1100T is,especially in minimum fps department.
I don't think PR team has the necessity to figure out the cause/reason.Quote:
Originally Posted by paulbagz
It is just that some people are interested in seeing how new scheduler/coding could help in improving BD's performance without any hardware changes.
I personally am interested in what the hell they did with all those transistors? As I understand it, the whole point of CMT was to create a very flexible design that was lite on resources to allow for massive parallel expansion. Seems to me that they just created the opposite instead. As a biomed with a minor in EE, I just find the circuitry to be intriguing
...where are the interlagos benches and products already? i mean i know cray got the first batch but...shouldn't they have made more by now?
I think the "major players" will be getting all shipments of Interlagos for a while. yields are too low right now to support sending server chips out to retail places for people to buy, especially with basically the same thing available for consumers already
Interlagos/valencia is normally scheduled for 11/11/11 but it is not an official date.Quote:
Originally Posted by god_43
current bulldozer chips are definitely broken, of that you can be sure. the transistor budget has not been spent as they intended.Quote:
Originally Posted by AliG
SourceQuote:
Originally Posted by Phoronix
In some cases the improvements are huge. Should be interesting to see how Bulldozer will turn out in the long run
any truth in this?
http://quinetiam.com/?p=119Quote:
After our own reader-base flagged the matter in comments last week, we began to investigate a rumour that there was a registry patch in the works that could offer a 40 per cent performance increase for AMD’s Bulldozer CPUs.
The issues brought up by the article are true, but I'm not sure if a simple registry patch will be enough to fix it.
Windows 8 employs a new scheduler to better optimize for Bulldozer, and from the developer preview, we can only see a few percent boost in performance at most.
It'd be great if this was true, but I highly doubt it will happen, otherwise AMD would have worked with Microsoft to get the patch out in time for the release of BD.
The increases are truly impressive, but not across the board for the AMD compiler. Furthermore, it may be feasible to recompile your applications when running on a server dedicated to a specific task, thus optimizing the application for the hardware it will be running on and get great performance. But you can't achieve the same thing when you're running Windows with a bunch of closed source, unoptimized games and applications. So as much as the potential performance of Bulldozer is great, it's still not utilized to the fullest due to the lack of OS and compiler (of current apps) support.Quote:
Originally Posted by BrowncoatGR
I still remember JF-AMD's last post, almost a month ago just before this crap was released:
We are used to this guy's arrogance. He knew about BD's perfomance, and he just played with the fanbois for months. In the end, you get what you deserve.Quote:
Originally Posted by JF-AMD
Funny thing is that people are still looking for magical fixes and other bullcrap. Just let it go guys, BD is a complete failure and you won't change that.
The obvious solution is to quit using closed source software that you can't recompile when you wish.Quote:
Originally Posted by The Coolest
It's official. All those 2012 doomsday theorists are AMD fans.
And in the end this isn't rally iteressting for the people here anyway... most common compiler for the windows platform is the MSVC. Its nice that you can gain some gain, but if you can't get them in 95% of the apps out there what is it worth it?Quote:
Originally Posted by BrowncoatGR
Intel on the other hand still managaes to get ~5-10% avg ipc increase each generation compared to the previous one in the same apps, so if you run the same apps you actually gain something and you don't have to hope that one dev decides to recompile there programm and also with the right compiler.
Im going to buy one just for cpu-z benching :)
AMD FX-8150 Multi-GPU Gameplay Performance Review
November 03
http://www.hardocp.com/article/2011/...mance_review/1
No suprise there, once you remove the gpu bottleneck (tri sli) BD only sees dust from SB.
Dreadful to say the least. Acts as an Athlon 64 single core. Pathetic. What were they thinking?Quote:
Originally Posted by cold2010
Another SLI benchmark set:
http://www.madshrimps.be/articles/ar...#axzz1ceA1SgT8
The BF3 result is especially interesting since the game was chosen by AMD to present BD as an "enthusiast CPU".
We all know the design is broken. It's just literally a matter of what Redbull78 said, "What were they thinking?". They must have known for at least a year that the design was broken. Having said that, I own an 8120 and it's not such a bad CPU.
Give the CPU some time to mature and we'll see how far it can stretch it's legs. :)
BD with different compilers in linux :
http://openbenchmarking.org/embed.ph...ha=830ac33&p=1http://openbenchmarking.org/embed.ph...ha=6fe0918&p=1
Link : AMD Bulldozer With GCC, Open64, LLVM/Clang Compilers
Published on November 02, 2011
Now compare to Sandy Bridge
http://openbenchmarking.org/embed.ph...ha=830ac33&p=1http://openbenchmarking.org/embed.ph...ha=6fe0918&p=1
Link : AMD FX-8150 Bulldozer On Ubuntu Linux
Published on October 24, 2011
Also See : Open64 Compiler Tuning On AMD Bulldozer FX-8150
Yeah, if gaming is your top priority Bulldozer with it's very weak cores is the worst buy for the price. And it's further exacerbated by the fact that most games today are compiled with ICC.
clock speeds are crap in that review, 4.6ghz on an FX-8100 would be a complete waste, and someone would rather disable some cores or go with a 6100 and overclock it higher. it wouldnt be enough to beat a 2500k, but it would bring the gap up way closer is some titles.Quote:
Originally Posted by cold2010
Waow, that's really weak :eek:Quote:
Originally Posted by cold2010
Marketing about a real gaming processor failed completly :shakes:
Seeing alot of talk about bd being buged and not running a lot of games, even bsoding on some titles.
looks like AMD is cutting 10% of its workforce,according to the Associated Press.
AMD to cut 10 pct of workforce as PC weak
maybe they can put that extra $10 million to good use?
The layoffs and other actions are expected to save $200 million in savings in 2012Quote:
Originally Posted by radaja
double post, and I only hit the button once...
yea it said,10m this quarter and 118m next year but........Quote:
Originally Posted by Stukov
but now it looks like they changed the article to show estimates of how many employees will go and the new estimate of 200m?
i sure hope JF-AMD doesn't get cut from AMD
Olivion: but this is multi-GPU, not single GPU. This was "clear"....
TDP limit is still on and thus its down clocking?
BD is a crap cpu, broken arhitecture, that's the only motiv.
Plain and simple! :down:
Steve Jobs proved that if you push people to their limits, you'll see greatness. I don't think AMD's problem is money, it's leadership. They have so much instability in their board and the CEO seems to be a rental position these days. I think all they need is a guy that's not afraid to get into their employees faces from time to time and tell them they aren't cutting it.Quote:
Originally Posted by radaja
Granted I'm way over simplifying things, as Jobs did have a tremendous ability to read people's emotions and be charismatic as well. With that being said, before he went off on the deep end of price gauging etc, he basically changed the entire computing industry with a team 6 of people. If AMD can find a guy like that they'll be competing with Intel again in less than a decade (yeah that seems like a while, but let's be realistic, this architecture sucks, it'll take at least 1 or 2 more to get back on the ipc playing field).
I see a lot of talk about windows 8 helping bd due to better thread managment, seems to be the prefered excuse atm, am just wondering if that will also benifit intel cpus? Probaly not the 2500 with no ht but will better thread managment not help ht enabled cpus like the 2600?
Unlike the bulldozer's case,Windows 7 already handles i7's SMT cores quite well.
How do we know that? Mabey win8 will give both amd and intel a boost?
I tested 4/4 and it isn't faster clock per clock, at least not in the games I tested it with... and to be honest the end user shouldn't be bothered with this fine tuning at all... Only thing that it allows you to do is to clock up higher and drop power consumption... But it will need to be at least at 5Ghz and beyond to give the raw power of the Intels a run for their money... and for that to be usable 24/7, Global Foundries has to make up for their crappy production process.... as they are responsible for the final blow to this new architectureQuote:
Originally Posted by Manicdan
The weak multi GPU game performance can be easily spotted in 3DMark06 and Vantage as the GPU tests are way too low. Even lower as a Thuban at 4Ghz ( FX8150at 4.6) Anyone ever wondered why we saw so many Heaven benchmarks ? Coz its 99% if not more GPU limited, if you even loose 100 points there on the score, something is terribly wrong...
Solution to this all : harsh price cuts and a better GF production to allow us to run these things over 5Ghz... All the other software recompiling and stuff is encouraging, but most of the image damage has been done... and later this month another thing is coming...
Also most reviewers are testing with ES CPUs that don't seem to clock as high as the retails, mine would run the full test suite at 4.9, though prime stability was at max 4.6-4.7Ghz...
Leeghoofd......"and later this month another thing is coming...", care to elaborate?
I guess Leeghoofd talked about SB-E launch coming in 10 days, the 14th :shrug:
other two reviews
http://www.xtremehardware.it/eng-rev...-201111146032/
http://www.xtremehardware.it/eng-rev...-201110126039/
maybe if you want to add
other two reviews
http://www.xtremehardware.it/eng-rev...-201111146032/
http://www.xtremehardware.it/eng-rev...-201110126039/
maybe if you want to add
How is that a review? It's an analysis of the architecture, they dont even have any form of performance test in the article...Quote:
Originally Posted by tom1
AMD confirms: Bulldozer FX only 1.2 instead of 2 billion transistors
Quote:
Originally Posted by Googlish
????
Huge ass die on GF's 32nm process is only 1,2 billion transistors? Llano has 1,45 billion transistors, 228 mm^2.
BS.
Why does so many people blame GF for bulldozers failures? Even if Bulldozer was at 5GHz+ it wouldn't touch 2600K territories. No company can produce bulldozer at those frequencies. And I doubt any company can make a huge chip with 2 (or 1.2?) billion transistors energy efficient at frequencies around 4GHz, compare with SB-E.
GPU part in Llano supposedly has very high transistor density.Quote:
Originally Posted by STaRGaZeR
Lol wat?
marketing doesn't even know how much transistors there own product has?
WTH is going on in AMD? :shocked:
Bulldozer :eek:Quote:
Originally Posted by Hornet331
I know you like to suppose a lot, but the official figures are ~2B transistors for the die and this is pure BS. That or AMD's PR department just hit another level of mediocrity. 1,2B on a process that is known to be more dense than the competition? ;);)Quote:
Originally Posted by informal
And you to... density is also dedicated by the logic used and not only by the process itself. Amd always had worse cache density on there cpus and suddenly they beat intel by a large marigin...Quote:
Originally Posted by STaRGaZeR
The SOI process technology is rather counter-productive for dense logic structures - certainly not as good as node-matching bulk process.
Official number has been corrected now,it's 1.2B and die size is 315mm^2. Bulldozer is server oriented chip and a lot of stuff was made with that target in mind. That's why it's somewhat larger die. Also as Hornet said, intel always had higher cache density versus Kx generation.Quote:
Originally Posted by STaRGaZeR
Where is that official correction?