AMD Bulldozer [official thread]

[smarac]

reborn .... tj zeljko .... ajde molim te zaobidji ovaj forum taman sam pobegao sa bencha zbog tebe i tebi slicnih koji su sklopili 2 kompa u zivotu i tripuju da znaju nesto ako koriste termine koje su videli na netu .....

Idi radi ili sta vec ..... ne zelim da se umaram preskacuci tvoje glupe postove, a koliko vidim i ostali su suvise zauzeti da tebi odgovaraju i da ti crtaju nesto umesto da se vodi neka normalna rasprava.

 

[solajam]

Ne treba niko da zaobilazi forum , bice monotono, jedan prica a ostali bleje .)

 

[reb0rn]

@drfedja
nisi me razumeo, mislim ako je na primer GPU razvoj brzi i sledece godine izadje 2x brzi, da on nece zahtevati duplo jaci procesor da bi mogao da iz sebe izvuce maximum vec ce u najgorem slucaju zahtevati nesto jaci CPU (frekvenciju, IPC, ili multitread engine igre, ili mozda cak brze baratanje sa memorijom/cash bas ono sto phenom II fali)

 

[voodoons]

reborn .... tj zeljko .... ajde molim te zaobidji ovaj forum taman sam pobegao sa bencha zbog tebe i tebi slicnih koji su sklopili 2 kompa u zivotu i tripuju da znaju nesto ako koriste termine koje su videli na netu .....

Idi radi ili sta vec ..... ne zelim da se umaram preskacuci tvoje glupe postove, a koliko vidim i ostali su suvise zauzeti da tebi odgovaraju i da ti crtaju nesto umesto da se vodi neka normalna rasprava.

A ja cu izgleda morati crtati..

Kome se ne svidja nacin na koji taj izrazava svoju misao i misljenje, moze slobodno koristiti integrisanu funkciju "Add this user to ignore list" i svi mirni i spokojni..

 

[BEGINER]

A ja cu izgleda morati crtati..

Kome se ne svidja nacin na koji taj izrazava svoju misao i misljenje, moze slobodno koristiti integrisanu funkciju "Add this user to ignore list" i svi mirni i spokojni..

Kratko i jasno. Jos jedna stvar koju bih dodao, ovo je forum, i svako ima pravo na svoje misljenje, bilo ono ispravno ili ne... Nekad je to misljenje argumentovano, nekad nije, moze da vam se svidi odnosno ne svidi, da probate da diskusijom promenite misljenje te osobe za koju smatrate da nije upravu ili da promenite vase misljenje ukoliko ta osoba dopre do vas , na kraju krajeva, kad bi svi isto mislili, forumi i diskusije ne bi ni postojali ;, U slucaju da nijedno od navedenih ne pije vodu, i ako ne mozete da se ne nervirate, padate u vatru, regaujete na prvu loptu i slicno, i postovi odredjene osobe vam konstantno smetaju, ima i za to reseneje, kao sto je Voodooons rekao, opcija Ignore je spas Za slucaj da neko ne zna kako kako to da odradi procedura je sledeca:


1. Kliknite levim klikom na nadimak osobe koju zelite da stavite na tu listu

2. Izbacice vam se (pop-up window) sa odredjenim opcijama, medju njima je i View Public Profile, koja je ujedno i prva opcija, klik na tu opciju

3. Bicete preusmereni na profil te osobe.

4. Pojavice vam se profil, i prvo i uocljivo veliki avata ili neka druga slikar te osobe, odnsosno podaci te osobe preko celog ekrana i odmah ispod njega u levom uglu videcete opcije: Send Message i User Lists

5. Kliknite levim klikom na User Lists, izbacice vam se opcije Add to friends list kao i opcija Add to Ignore List

6. Klik na opciju Add to Ignore List, nakon toga kratki YES da potvrdite, i to je to, vise necete videti postove nezeljene osobe.

Znam da ovo manje vise svi znate, i bas zbog toga sto svi znate, nije me mrzelo opet da napisem ovo, nakon ovog mog i posta voodoons-a, svaki sledeci post koji bude slican kao sto je ciitirani post u poruci voodoons-a, ce momentalno biti sankcionisan udaljavanjem ili ti banom sa foruma u roku od 7 dana zbog krsenja pravila PCAXE foruma

 

[Diabolique]

Ne ložite se bezveze. Nema šanse da će sledeća generacija grafika biti brža više od 30% u odnosu na predhodnika, što znači da će overklokovan SB biti dovoljan za jednu kartu.
Sa druge strane postavlja se pitanje koliko ljudi ima 580 ili 6990 i koliko će kupiti najjači model sledeće generacije.

 

[PERUN]

Da se vratimo na temu dr Fedja je pominjao za septembar "paper launch" Bulldozera.Znaci li to da ce na tu biti predstavljen 100% onakvim kakav je a ne kako ga je Obrovsky predstavio,ili ce i tu biti znak pitanja kakve su mu mogucnosti u svim aspektima gaming,rendering itd...Ne bih sad kupovao 955BE a da posle moram da ga prodajem po nizoj ceni izdrzao bih nekako do decembra na ovom starom kekalu od laptopa,posto sam danas definitivno odustao od intela,ove kvalitetnije ploce z68 su dosta skuplje od am3+ ploca te je amd opet primamljiviji.

 

[reb0rn]

Pa kao ga predstave realno ili kao paper launch imat ces detaljan prikaz procesora.... jer ce svi vodeci review sajtovi dobiti svoj primerak koji naravno nije od papira

 

[PERUN]

Lepo,lepo samo da znam dal je dostojan za gaming

 

[solajam]

shta ce ti BD za gejming ima se AMD 955BE koji trenutno ispunjava sve zahteve a nije skup .)
Ako izmedju 955BE i SB 2500/2600K procesora nema neke drasticne razlike u gejmingu zashto bi to bio slucaj sa BD ???

A ona pricha zalog za buducnost, pa kad dodje ta buducnost bice sve slabasho i sadasnji SB i BD koji tek treba da se pojavi...

 

[PERUN]

Nemam ti ja pare za menjanje kompa svakih godinu dana,hocu da uzmem nesto sto je tek izaslo a ne nesto sto uskoro i prestaje da se proizvodi mislim na 955BE.

 

[Bogdan901]

Obrovsky:


The kids cry in all forums, because the Chinese results show B2 revision, what can I say for two months - Bulldozer performance is laughable, and IPC is worse than on Phenom II. That BD is better than I showed you and now the Chinese, believe it or slowly Flanker - AMD's biggest fan in the world.

He left only a dimwit - Informal on XS forums, everyone's already given up, have it - finally! It seems illogical to you, featuring a CPU with a worse performance than the previous chip? - BD is not multi-thread, worse than PII X6 1100T is about 20 percent more efficient ... but only in massive multi-thread applications. Elsewhere in the ass gets brutally fcked... somewhere is literally pathetic performance (DosBox, SuperPi, xvid conversion, photoshop). Average, but overcomes the aforementioned Thuban, the mission is fulfilled ...

But now as to why this is such a misery. Here we go into history, until 2007. It has been shown Phenom architecture Stars, at the time but now the other team worked on the Bulldozer and had him go in 2009. Like rival Core i7 (Nehalem). But things screwed up with Phenom and AMD had to devote lots of resources to the new revision, which took them 5 months. Now, remember that AMD is not a company with an unlimited budget, rather the contrary - is in the mess for a long time. There is not enough money for many projects mismatching ... Phenom II, Radeon, Bulldozer, APU, servers ... something has to go away - Bulldozer.

If this went on 45nm chip at a time when he had, he would be a great match for the Core i7-920, i7-965XE ... and more. against them as was planned, the plans are really nothing has changed, only the node is smaller - 32nm, but the proposal is still the same - now inadequate. Intel AMD meanwhile ran for two years in advance.That said, Sandy Bridge, on which BD does - especially in single thread performance, gaming performance ... can compete with him except in CINEBENCH and similar applications for eight cores.

If some morons (Stach) will want you to assert that it really does not matter, so that's really the Institute of denial. Why do you think Intel puts the highest emphasis on single thread performance (extreme turbo with SB for one to two threads)? - Because 90 percent of current applications (which use the most) benefit from the exercise of one or two cores. Powerful multi-thread use the monsters really just pros, not the gamers and ordinary users, but count yourself how much your home applications using cores. It will be better performance on two quick thread on SB or eight cores BD slow? - The answer is clear ...

Why is BD slow and slower than the Phenom II somewhere? is an entirely new, untested architecture tested the ice on simulations and models, it also created a different team than Phenom ... formed simultaneously from the beginning. Thus, the team could not know how you will stand against Phenom of 2011 ... if they went out in 2009 would be more powerful than anything on the market, including Core i7 ... Unfortunately, now is not enough.

Indeed, for AMD it is no problem. Radeon HD 6800 is slower than HD 5800, the first Phenom were often slower than K8 ... BD and so will just a bit faster than Thuban ... so what, it already is.

PS. Or do you think that if it was such a bomb, so AMD is no longer fails? Or show up at various events performance? Not discharged the traditional forum-flame-war on the selected sites, forums and ministers? Yeah, reconciled with what I showed you a Chinese show, the real power BD shit ... the price 300 USD is only wet dream of some farmers in zfetovaných AMD will have to go down well ... perhaps even lower than I've said some time ago ... But you will see.

PS. Honestly, there only a miracle will help AMD. I am not saying that the performance of my B2 and the Chinese is the final, but I think so. If the performance did not change B0 to B2, which are separated by a chip factory 4 months, I do not believe in any miraculous secret B2 revision, or further review I do not expect products to reviewers go slowly, putting the already counted the days ...

 

[Steamroller]

Dobro što nisam kupio AM3+ ploču.

 

[Bogdan901]

AMD Insiders Speak Out: BAPCo Exit is An Excuse for Bulldozer

The Beef: Why AMD Attacked BAPCo Now - Bulldozer CPU Performance
If you are wondering what lies in the timing of this announcement, our sources were pretty clear.

"Do you know why AMD chose now to go after SYSmark now?”

We expected that the source would say Llano APU performance etc., but no: "Because Bulldozer performance is sub-par. Because after such a long wait Bulldozer will not be a world beater. They knew that for another generation, we would come up way short on benchmarks like SYSmark. Removing ourselves from BapCo was the best way to make the benchmark look like it's no longer valid in the hope that prospective or existing customers stop using it."


Now, BSN* had received scores from B0 and B1 silicon and we decided not to run the story disclosing exact performance details since they plainly... were far below what we expected. After talking to journalists from other publications and developers that had access to engineering silicon, everyone was lead to believe that Bulldozer has a lot of "performance locks" which were set in place to prevent "premature benchmarculation" i.e. leaks.

"Bulldozer is going to disappoint people because we did not get the resources to build a great CPU, and it's not that we needed billions of dollars to make it a leader. We needed investment in people, tools and technology."
As every competitive spirit, one of our sources went on to confirm the quote above and expressed concern that "Bulldozer’s client performance is not going to be enough to deliver growth in the enterprise and in my view, that is the biggest driver behind our current strategy."

When asked about core performance, surprising information was that a Bulldozer core versus the existing cores in Llano will result in minimal improvements overall. Our sources went on to say that the launch of Llano clearly shows what is the current and future strategy - downplay CPU performance every chance you get. Everything has to revolve around the GPU.

Comments continued "As soon as you have analyzed the Bulldozer core, you're quickly going to find out that it disappoints at SYSmark. It won't matter, because by that time I am sure the current marketing strategy will be changed again when we have our new CEO in place."

"What We Want from AMD"
As the article draws to a close, we aren't going to comment on the following statement made by yet another source:

"There are more people like me at AMD that want to go and fight from an engineering perspective in the same spirit of the old Jerry Sanders days. We don't want an AMD that is going off and writing blogs about transparency and fluff, while engineers are frustrated. Engineers are not getting the resources they need, we're not getting the time and the investment we to build and optimize new architectures.



Instead, they feed marketing fluff and kool-aid, and all these self-centered blogs - instead of focusing on doing what we know how to do best. Fight, survive and challenge against all odds.



Thankfully, we'll have a new CEO soon and hopefully all this strategy nonsense will be out the door."

A Message to the new CEO
Furthermore, this is a personalized message to AMD's new CEO:

"Many of our best engineers are leaving to join companies like Apple, Freescale, Qualcomm, Samsung even NVIDIA and Intel - people that thought they would never work for another company.



Look at AMD's history - how we beat Intel Pentium III, which was based on one of the best CPU architectures of all time - the Pentium Pro - and we beat the them with the K7 (original Athlon). Look at K8, 64-bit innovation and the integrated memory controller. Look at how ATI came back from the brink and beat by NVIDIA soundly with R300 and took performance leadership across the board.



We created the netbook in OLPC project with Nicholas Negroponte when everybody was laughing at him and then gave the whole market to Intel. We were shipping millions of graphics chips for cellphones and were in TVs long before today’s leaders. We know how to win."


Conclusion
At the end of our conversation, we asked our main source what does he plan to do, was this rant his farewell with AMD or not. The answer was plain and simple: "I was here before they came, I am here with them now and I am going to be here well after they’re gone. My heart and soul, and more importantly my legacy is with this company and will continue to be."

Time will tell, but as the proverb from my former life says - all stories come true.

 

[Kyau]

Uff koji FUD ! Plus OBR vec poznat kao lazov (zbog toga i banovan sa XS), pa on i nema Bulldozer, lazirao sve rezultate ! Fud more !

 

[drfedja]

Da bi FX8 bio jednak Phenomu II X6 po performansama, morao bi da ima oko 5% manji IPC od Phenoma, po modulu, a po jezgru 15-20%. To onda nikako ne bi opravdavalo cenu od 300+ $. Da ne pricam o FX6, koji bi u tom slucaju bio jednak Phenomu II X4, to sve onda ne bi imalo nikakvog smisla, a svo istrazivanje i razvoj bi predstavljalo cisto trosenje resursa od vise milijardi $. Mislim da oni nemaju takav luksuz.

 

[Diabolique]

AMD rep John Fruehe

Q. I saw a benchmark on xyz website. Is that how bulldozer will perform?
A. No. Nothing posted before launch will be representative of actual performance. To get actual performance, you need:

Final production silicon
Final processor microcode
Final system BIOS
Final OS optimizaitons
Final drivers
An app compiled with the latest flags
A person who understands the app and configures the test properly

Without these things (and there are probably more), you cannot get an accurate benchmark. Any extrapolation of a crappy benchmark gives you a crappy estimate of actual performance. Period.

Oh, and many of the benchmarks that you see were probably not run, those are just charts made in excel. It's really easy to make a chart in excel - what do you want, bulldozer faster by 3716%? Intel faster by 293%? Sure, I can do either one in 10 seconds. Why are you arguing about benchmarks that probably aren't real?

Q. When are you launching? Why don't you release the date?
A. When we launch, we launch. I will not comment on dates, I will not comment on schedules. We do not release dates prior to launch, at most we give quarter granularity. Giving the date out will stall demand. We have a business to run. While you might think that it will make your life easier to not have to guess, the reality is that we have a business to run and the minute you let the date out, sales stall. For everything. The cost impact of announcing the launch date is always bigger and drives these decisions.

Q. What are the prices?
A. Look at the above question and you have your answer.

Q. Why don't you release benchmarks before launch? You could steal so much business away from the other guy?
A. Again, releasing benchmarks before launch will simply stall sales. Believe me, if the competition thinks they are out of position, will they just sit back and say "oh well" or will they react? Handing them benchmarks is simply giving them time to form a strategy. I am not in the business of helping them, they are on their own on this one.

Q. Will bulldozer be faster than....?
A. I don't need to finish the question, please read above.

Q. I read on xyz site that you were launching on xxxxx?
A. Yeah, and I read on another site that elvis was still alive. The reason I don't comment on date rumors is that there are a limited number of days in the quarter. Once you say no to some, and suddenly say "no comment" or don't answer that one, immediately everyone thinks that is the date. So, no matter how crazy it sounds, you can't answer any of them.

Q. I saw someone selling bulldozer parts online, that must mean the launch is happening, right?
A. No. First off, many of the people that are advertising parts for sale ahead of time do not have parts in hand. Buyer beware. If they are selling engineering samples, we will take care of that. Occasionally parts are loaded into disti databases, and if the flag is set wrong, it can flow through EDI to a partner's database and show up online. That does not mean parts are available. Oh, and sometimes distis use planned pricing and part numbers as placeholders, so don't believe what you see. Nobody is allowed to advertise parts ahead of launch.

Q. I know someone who knows someone who has a cousin who lives next to...?
A. Ask yourself this. Would you take investment advice from someone like that? If not, put a big grain of salt into those conversations.

Q. Why haven't you guys launched?
A. The key to a successful product is lauching it once you are ready. When you are trying to put hundreds of millions of transistors onto a piece of sand about the size of your thumbnail, there is a lot going on. It's a pretty complicated busines.

Q. Why don't you comment on client stuff? Why don't you get a client guy to post on here?
A. I am in the server world, I comment on my products. I will not comment on their products, because I don't want them commenting on mine. The few times I have ventured into the client world, I have caught flack for my comments. It is not worth the trouble. Based on the filth that seeps into my inbox from a very small segment of the enthusiast world, I can understand why the client people might not want to make their names known. When you are attacked in a vicious way for no reason, and you are doing this on your own time, you start to ask "why bother"?

Q. I see you have on server, so we will see that on client too?
A. I can't say for sure. We both productize different features based on our customer needs. Wait for the launch to see for sure.

Q. Why aren't you doing an unlocked Opteron so I can do 2P overclock?
A. First the market for 2P client systems is tiny, and getting smaller every day. While you would like to buy one, there just aren't enough people to do that. Second, I want to focus my time on servers, client overclocking will not be supported because it takes efforts away from making virtualization and cloud computing better.

Q. If AMD does beat intel in xyz it is the end of the road
A. OK, this is not a question, but enough people make statements like this. AMD is a 40 year old company, not likely.

Q. Is IPC higher on bulldozer? All I care about is IPC.
A. IPC is simply a measurement. What if IPC was 2X what it is today, but clock speed was 500MHz. Is that what you want? You are getting double IPC, right? IPC is only one measure. The people that are telling you IPC is the only thing that matters have an agenda. Taking only one measurement, out of context, is like trying to say that a person's weight is all that matters. I weigh 195. Does that make me fat? Does that make me skinny? It is impossible to say unless you know my height. IPC is like weight - it tells you something in context to other factors, but is meaningless on its own.

Q. What about single threaded performance?
A. See above. Also, if all you care about is single threaded performance, might I recommend a lovely, inexpensive single core processor for your system?

__________________
John Fruehe is the Director of Product Marketing for Server, Embedded and FireStream products at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites, and references to third party trademarks, are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third party endorsement of AMD or any of its products is implied.

http://blogs.amd.com/work/author/jfruehe/

 

[drfedja]

Pa lepo je rekao. Vecina ovih nazovi "leakova" su cist bullshit i sigurno ne daju realnu sliku o tome sta ce biti.

 

[Vule.ue]

Sve sto je ovaj lik rekao je treba nam jos vremena i nista vise.

 

[drfedja]

Drugim recima, BD moze da bude lansiran ovog meseca, a moze da bude lansiran i u Decembru.

 

[reb0rn]

Q. Is IPC higher on bulldozer? All I care about is IPC.
A. IPC is simply a measurement. What if IPC was 2X what it is today, but clock speed was 500MHz. Is that what you want? You are getting double IPC, right? IPC is only one measure. The people that are telling you IPC is the only thing that matters have an agenda. Taking only one measurement, out of context, is like trying to say that a person's weight is all that matters. I weigh 195. Does that make me fat? Does that make me skinny? It is impossible to say unless you know my height. IPC is like weight - it tells you something in context to other factors, but is meaningless on its own.

Ovo je za mene los znak.... ako ti je IPC isti (ovaj tekst meni naglasava da je isti barem ga ja tako tumacim) onda da bi to nadoknadio treba ti dosta veca frekvencija od intel-a (za sta je potreban napredan proizvodni proces a tu znamo koliko je intel ispred amd-a), ok vise jezgara ali bez optimizovanog software-a desktop korisnici nemaju nista.... server i odredjena primena je druga prica gde se koristi mulitreaded software
E sad verovat da ce se Buldozer klokovat bez problema (vazduh) na 5Ghz+ tesko, ipak je to relativno velik CPU die i "samo" 32nm proizvodnja u narednih 18 meseci ako ne i duze

 

[Beagle]

^
Pa sta ako je veliki CPU die? Phenom X6 je po svim danasnjim merilima ogroman, pa se opet klokuje ko blesav.

I covek je u pravu, IPC je samo jedan pokazatelj. Nek ostane identican IPC (a nema sanse, moze da bidne samo veci), osnovni radni takt im je osetno veci od aktuelnih Phenoma + Turbo + overklok i dobijas sigurno brzi procesor.

Jos se nije desilo u istoriji da novi procesor bude sporiji od prethodnog, nemojte bez veze da tupite.

 

[drfedja]

Ovo je za mene los znak.... ako ti je IPC isti (ovaj tekst meni naglasava da je isti barem ga ja tako tumacim) onda da bi to nadoknadio treba ti dosta veca frekvencija od intel-a (za sta je potreban napredan proizvodni proces a tu znamo koliko je intel ispred amd-a), ok vise jezgara ali bez optimizovanog software-a desktop korisnici nemaju nista.... server i odredjena primena je druga prica gde se koristi mulitreaded software
E sad verovat da ce se Buldozer klokovat bez problema (vazduh) na 5Ghz+ tesko, ipak je to relativno velik CPU die i "samo" 32nm proizvodnja u narednih 18 meseci ako ne i duze

Ti si nepopravljiv.

Ne, covek nije rekao da je IPC isti ili manji, rekao je isto ono sto ja ovde trubim vec ko zna koliko. IPC je bitan, ali je bitan balans izmedju IPC-a i frekvencije. Sta ti znaci 2x veci IPC ako procesor ne moze da ti radi na vise od 500 MHz. Ako ti treba veliki IPC uzmi Elbrus CPU taj ima IPC 20.
Da li je najbrzi zbog tako visokog IPCa? Pa naravno da nije.
Sto se tice povrsine cipa, jeste relativno velika, ali to ne znaci nista za klok.
Kriticne povrsine kao sto su moduli su relativno male. Jedan modul = jednom SB jezgru po povrsini.
A klok zavisi od mnogo faktora.
IPC != performanse, to zapamti jednom za svagda. IPC je relativan i zavisi od softvera i hardvera na kojem se izvrsava. Softver moze da ima odredje ILP (instrukcijski paralelizam), a IPC moze biti manji ili veci u zavisnosti od ILP-a i samog hardvera.
Da li treba da ti crtam funkcije koje demonstriraju koliko ti treba resursa za drasticno povecanje IPCa?
Za 50% povecanja IPC-a u odnosu na 10h, trebalo bi ti ogromno jezgro. Znaci, ne cpu die, nego bas jezgro, sto nikako nije energetski efikasno i nema potencijal za klok.
I mislim da u JF-ovim izjavama ne treba traziti nikakvo "skriveno" znacenje i tumacenje performansi BD-a.
Stvar je tu vrlo prosta.

^
Pa sta ako je veliki CPU die? Phenom X6 je po svim danasnjim merilima ogroman, pa se opet klokuje ko blesav.

Dzaba mu objasnjavamo. Covek ocigledno ne shvata.

I covek je u pravu, IPC je samo jedan pokazatelj. Nek ostane identican IPC (a nema sanse, moze da bidne samo veci), osnovni radni takt im je osetno veci od aktuelnih Phenoma + Turbo + overklok i dobijas sigurno brzi procesor.

Dovoljno je da ima u proseku 15% veci IPC na nivou modula i CMT skaliranje od 70-90% i bice brzi najmanje 35% od Phenoma II X6. To je jasna racunica.

Veci IPC ne znaci vece performanse, bas kao sto i veca frekvencija ne znaci vece performanse. Throughput je takodje mera. Npr, imas Concorde koji leti brzinom od 2000 Km/h i prevozi 150 putnika i imas Boeing 747 koji prevozi 400 putnika i leti brzinom od 900 Km/h. Koji ce vise putnika da preveze po satu ? Pa naravno, 747. E pa isto je i sa brojem jezgara - threadova. Ako kazu da Intrlagos ima 50% veci throughput, to znaci da moze 50% vise posla da odradi u jedinici vremena od MC.

Jos se nije desilo u istoriji da novi procesor bude sporiji od prethodnog, nemojte bez veze da tupite.

Pa cak ni Prescott nije bio sporiji od P4 NW. Bili su tu negde....

 

[reb0rn]

Ne vec sam mislio da je njegov stav po pitanju IPC-a dosta "ladan" (za mene), a ostala prica je cisto mlacenje prazne slame zato sam samo ovaj deo izvodjio... jer u sustini ama bas nista drugo nije rekao

 

[drfedja]

Ne vec sam mislio da je njegov stav po pitanju IPC-a dosta "ladan" (za mene), a ostala prica je cisto mlacenje prazne slame zato sam samo ovaj deo izvodjio... jer u sustini ama bas nista drugo nije rekao

Da, njegov stav jeste takav, ali i od svakoga ko se malo vise razume u problematiku. Licno sam se zezao i pravio one excell grafikone i shvatio da njima zapravo ne treba IPC veci od 15% po modulu, sto je slozices se nije neki napredak, ali je sasvim dovoljno da to radi dobro. Uostalom, to kako ce raditi CPU ne zavisi samo od IPC-a. Evo npr, Llano ima 6% veci IPC od Phenoma II, pa jel brzi? Pa niiijeee. Zasto? Zato sto nema L3 cache, zato jer deli RAM sa grafikom pa je latencija veca. Llano je dobar dizajn, ali nije pravljen za performanse, nego za sasvim drugu pricu. Llano je nesto malo brzi od Athlona II X4 klok za klok. Ono sto je bitnije u celoj prici je kolika je razlika u frekvencijama.
S' druge strane BD je predvidjen za visoke frekvencije. FX 8170 koliko vidim ce standardno raditi na 3.9 GHz + turbo od 500-600 MHz sa 8 jezgara. Intelu sa Core arhitekturom to nece poci za rukom ni u 22 nm sa tim brojem jezgara. Razmisli malo o tome.

Cela prica oko IPC-a je prenaduvani bs od strane intel fanboy-eva. IPC jeste donekle bitan, ali nije najbitniji. Mnogo je bitniji balans izmedju IPC-a i frekvencije.
Zbog cega je i7 brzi od Phenoma? Pa zbog toga sto ima dobar balans frekvencije i IPC-a. Opet s druge strane, u 45nm AMD sa Thubanom ima sasvim solidne multithread performanse. Ima frekvenciju i broj jezgara. Uporedi Bloomfielda sa 4 jezgra i Thubana sa 6, pa ces videti o cemu pricam. Bloomfield jeste mozda brzi u Sysmarku, ali u vecini rendering testova nije. Moze se reci da su performanse tu negde. Disipacija takodje. Zasto? Pa zato sto manja jezgra zauzimaju isti termalni okvir kao i veca jezgra. Cilj BD-a je da zauzme manji termalni okvir sa vise jezgara. Npr. Interlagos sa 16 jezgara na 2.5 GHz ima TDP 115W. Xeon E7 ima 130W na 2.4 GHz sa 10 jezgara. 10 jezgara tesko da ima veci throughput od 16.