AMD CPU speculation... and expert conjecture

Page 233 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

Ranth

Honorable
May 3, 2012
144
0
10,680

If you would use your brain... just because Elric (Who seems way too happy about packaging) is braindead sometimes doesn't mean you have to be the same.
The fx 9590 is an OEM part....... The only concern they have is getting rid of the packaging, if you're getting like a 100 of those cpu's do you care even the slightest bit about the quality of the box??! :fou:

Also why does AMD's extreme series have to come in something more fancy than Intel's extremes...?
I wanted to do something horrible to Elric after I saw that video, he spend 5 freaking minutes ranting about how bad the packaging is, making every freaking mindless fanboy add "AMD lack of awesome packaging" to their list of "Why AMD should die".... why is it people won't use their brains.... :fou: :fou: :fou:
Sorry about my wording but I have a hard time when "big" people lack the ability to think and feed the "graving souls" with misinformation. :pfff: :(
 


The packaging could have been better with the ones sold via Newegg or other sites in their combos.
 

Ranth

Honorable
May 3, 2012
144
0
10,680


Well, yea. But as far as I know AMD never intended to have this released to people due to the motherboard issue.
 

GOM3RPLY3R

Honorable
Mar 16, 2013
658
0
11,010


Even so, do you know how easy it could've been just to drop it and break it, or step on it by accident? It's just two pieces of plastic in a small cardboard box. At least if they had Styrofoam in a slightly bigger box, that'd be safer. It's really just the safety of it. If I had that box delivered broken, I would take either AMD, the shipping company, or both to court if they don't replace it.
 


http://www.youtube.com/watch?feature=player_detailpage&v=0GfvHwp-6NI&t=6 Nuff said.
 

juanrga

Distinguished
BANNED
Mar 19, 2013
5,278
0
17,790


I have been revising jaguar performance and now believe that Feldman's claim that Steamroller cores are 2x faster than Jaguar cores did mean that both have more or less the same performance per clock but Steamroller goes up to 4GHz (approx.) whereas Jaguar only up to 2.0GHz.

Using the performance equivalence between 4 jaguar cores @ 2 GHz and i3-SB, and multiplying by 2x the resulting performance for 4 Steamroller cores @ 4 GHz coincides very well with the performance predicted by assuming a 20% increase from A10-6800k.

Therefore the two known ways to predict SR performance

jaguar --> SR
PD --> SR

coincide.
 

hcl123

Honorable
Mar 18, 2013
425
0
10,780


Not so sure...

Hafijur abhorrent intellectual dishonesty, persistence, and tolerance to some abuse... makes him of the *professional* quality...

He may had been assigned to this forum... and they should have been paying him well... if don't (most of cases) just makes hafijur *another one* of the most imbecile examples of human kind lol... not a troll, goes beyond and above lol...

 

cowboy44mag

Guest
Jan 24, 2013
315
0
10,810


+1 very possible. AMD would want stock levels of Richland and Trinity APUs to decline before releasing something better.

I also wonder if the delay isn't "software based" as well. By delaying a few months the HSA Foundation may get some more applications developed that utilize HSA before the launch of Kaveri. That way Kaveri can actually be benchmarked using applications with full HSA support and showcase its strong points.
 


I think he is referring to 8350rocks. It took our magic to remove Hajifur of his inner troll.
 

GOM3RPLY3R

Honorable
Mar 16, 2013
658
0
11,010


Yes, I am. I shorten his name so the 8350 does not rock. (^_^) Also, I posted my rig on the threads, and someone is trying to convince me to get a 4770k over a 3930k.

へ   へ
の   の
  も
  へ
 

etayorius

Honorable
Jan 17, 2013
331
1
10,780


I am still wondering why there are so many Fanboys and Intel Trolls in this thread, they just can´t keep their hands off AMD, because they have some love-hate relationship with AMD i guess.

It is cool that this guy is buying an Intel 3930k, but seriously the rest in here don´t care... but i seen this guy trying so hard to make it clear how superior and cheap that CPU is compared to any AMD CPU.

I do not visit Intel threads at any given time in other tech and forum sites, simply because i do not care... even less i would go to such threads and tell the Intel dudes i am buying "X" AMD CPU, they would give a crap and will probably rage that how dare me AMD fanboy to set foot in ther lands. Clearly this dude is either trolling or getting payed to keep his constant "buying the 3930k" because is so cheap and superior, geesh with this people.

Or maybe GOM3RPLY3R was expecting everyone in this Thread to applaud him like Seals while he tell us how superior the 3930k to the FX8350 while he gave his absolute support to Hafijur in pretty much everything he said prior to his tone down against AMD.

Now please, News, AMD, FX and SteamRoller? everything else just bogs down this thread, so stop it... you don`t like? cool, tons of Intel threads out there, buy your 3930k be happy and be done with it.

 


I admit GOM3R has to stop showing off his 3930K that he does not even own yet. He is not giving full support to Hafijur, telling people that they should stop going back and fourth with trolling. However, this is an AMD thread, not a hijack this with Intel thread.
 


Making those kind of remarks and trying to "lvl up" the forums contradict themselves. You remind me of what happened with Fez 2. A slow news day don't give you the right to trash someone/something to make it news worthy. What are you? 10 years old? IF you want a role model to follow on how to "lvl up" a place, go read what palladin has contributed, or MUEngy, gamerk, randomizer and other good Vets on the forums. Hell, even Baron gave good speeches from time to time without resorting to that type of offensive posting (IIRC).

I'm not in a position to tell you to go or not, but given that remark, a good vacation should be in place for you.

If you don't plan on growing up anytime soon, then I'd like to see you gone and come back when you're more mature.

Anyway... Slow or non existent news keeps us on edge I guess... The only place where I get new info is here actually, lol.

Cheers!
 

GOM3RPLY3R

Honorable
Mar 16, 2013
658
0
11,010


I apologize. I really only follow this thread anymore, as really, I love everyone (except the snotties) on here, and this is the only thread that I'm really following. Making those remarks was bad, I guess. I do, however, have the right to give a small comment like that every now and then, since once you went on a spree with someone for half a page about a UK War. >.>

I apologize to everyone again.
 

cowboy44mag

Guest
Jan 24, 2013
315
0
10,810


Slow news / No news is a b*tch, everyone drifts off target:p
 

BeastLeeX

Distinguished
Dec 13, 2011
431
0
18,810


I guess instead of posting non-sense we should all just relax and play some games on our "gaming computers" right?

 

griptwister

Distinguished
Oct 7, 2012
1,437
0
19,460
Lol, we're communist? I'll give you a better analogy. You, Hafijur and various others are the Westboro of this site: Judging users and telling them what they're doing wrong because they chose or are choosing to buy AMD instead of Intel, feeding these users lies like, their game play and CPU performance is damned!

Us and other Pro-AMD users are simply exposing the corruption behind Intel and this FanBoy mentality. (Like Anonymous with out hax) Exposing the real details, only to prove those points invalid. We find Intel does have better performance. But You and Hafijur are straight making up crap that's not stated in actual text (referring to the butchering of the bible that Westboro does).
 
That article is not a "few years old", it's published date is September 27, 2012, so almost one year old. They used what they knew how to use and after reading through it they're relying on someone else to give them technical information. The difference between CISC and RISC is not the size of the instruction set but the complexity of those instructions. Using small, less complex and more predictive instructions tend to result in smaller sets but not always. The goal of a RISC design is that each instruction takes exactly one cycle to execute and to avoid complex memory access modes in favor of simple flat design's.

Anyhow, GCC is an amazing compiler, especially the newer versions. It's the de'facto standard in the Enterprise and HPC world, where people actually know what their doing. The performance penalty their seeing is a result of the MiniGW environment not the compiler. GCC expects a fully POSIX compliant environment to generate code for, if that doesn't exist then compatibility libraries have to be introduced to smooth the code over. That in turn introduces a huge performance penalty. The newer versions do a much better job of compiling on Windows, but I still wouldn't use GCC on anything not POSIX.

The whole "ICC R EVILZ" thing has been gotten over for the most part. They are now forced to use SSE2 instructions on non-Intel by default which is a vast improvement over x87. The difference from SSE2 to SSE4/AVX isn't nearly as large. Most of their "tricks" are minor and while they do impact performance, not nearly on the same level as previously suspected. Mostly Intel CPU's + ICC just get preferential treatment with memory access and loop detection which is something we can all live with.
 

jdwii

Splendid


darn i guess that leaves me out
 

jdwii

Splendid


I own a 1100T clocked at 3.9Ghz and this CPU was released in DEC 2010 so its going on 3 years i can still play all my games just fine on it and honestly i feel like it aged better then other processors in that price range 3 year's ago, with games using more cores i feel like it will actually game better then a Intel processor from 2010 at least the ones priced at 200(i only paid 180$).

Now i do play on dolphin emulator and that emulator demands Performance per clock and high clock rates but only uses 2 cores and my phenom holds pretty well not as well as an Intel one granted but that is really the only time i would want a Intel machine. Seems odd but the new games seem to play better. My 6950HD seems to be doing OK at 1080P as well my machine handled crysis 3 at Mid-High settings at 1080P.

My friend games on my old 6850hd and his own phenom 955 x4 that machine is doing ok at 720P(his TV). Either way Amd is good and they age better it seems as well. Try a Atom from when bobcat came out- try a semperon compared to a celeron from back in the day, or a Phenom II x4 vs a I3 during its time frame. now to say that a 8350fx vs a I5 will be the same story would be a fallacy, but we are seeing patterns.
 

mayankleoboy1

Distinguished
Aug 11, 2010
2,497
0
19,810


Now Clang is also being extensively used, mainly because of the superior, out-of-box debugging capability. Also, i am not sure if the HPC crowd uses GCC this extensively, but yes, GCC is still the de facto standard.

The whole "ICC R EVILZ" thing has been gotten over for the most part. They are now forced to use SSE2 instructions on non-Intel by default which is a vast improvement over x87.Most of their "tricks" are minor and while they do impact performance, not nearly on the same level as previously suspected. Mostly Intel CPU's + ICC just get preferential treatment with memory access and loop detection which is something we can all live with.

Heh, tell this to the AMD fanbois, baying for Intels blood. Maybe they will listen more to a Moderator.


The difference from SSE2 to SSE4/AVX isn't nearly as large.

From what i see on Phoronix benchmarks, almost all programs get maximum speed from SSE3. After that,you go any higher, sometimes you get a regression.
 


Doubt it. HSA can't really be "implemented" in code; its handled entirely within the domain of the Windows scheduler.

Look, HSA is basically a bunch more cores, optimized for FP heavy, parallel workloads, within the same CPU context. Because of this, its up to the scheduler to put the right work on the right cores. Really isn't anything you can do in code to enhance this. (Unlike, say, OpenCL, where you do have control over what work gets offloaded to the GPU).

Hence my greatest fear of HSA: Performance loss due to the wrong tasks getting loaded on the wrong processor unit. Especially since, as we recall, the Windows scheduler wasn't updated in time for BD/PD...
 

8350rocks

Distinguished
@gamerk316:

Why do I get the feeling that Linux benchmarks will, once again, be dramatically faster for AMD architecture when all the dust settles?

Oh, that's right...the Linux kernel is light years ahead of Windows kernel...
 
Status
Not open for further replies.