AMD: Zen will offer 40% faster performance per clock than Carrizo

gamerk2 said:
My concern is this: While IPC is probably going to be 20-30% higher for the typical use case, there has been very little mention of the other half of performance: Clock speed. I have a feeling we might see a clock speed reduction compared to BD.

Point being, if the IPC improvements get you to Haswell level performance, but you then reduce the clock, you fall back to SB/IB level, which is basically non-competitive.
Click to expand...

griff30 said:
You think a drop in die size is going to decrease the clock speed?
Click to expand...


Clock speed has less to do with performance because of the base design of Zen, going wider and doing more per clock is much more important that clock speed. If they are able to keep their chips around the 3.5 to 4 ghz range and improve their IPC they are good to go.
 
razor1 said:
Do you know when those extensions are used?

Tell me and then we can talk about whats relevant and whats not. And also if the software doesn't take advantage of them, its irrelevant as they won't take advantage of them in Intel's case either.

This is probably the the 20th time go read instead of posting stuff you don't know about.
Click to expand...

You have some balls posting this while this is from the article itself
Bulldozer’s improved SSE performance (above) and AVX support (below) may help the chip in some corner cases, but at least some AVX-enabled benchmarks, like the Kribi 3D tests available at inartis.com are actually slower on Bulldozer when AVX is used than they are otherwise. It’s not clear if this is because Bulldozer’s AVX implementation is narrower than Intel’s, or because the chip’s SSE capabilities make that instruction set a better fit. Similarly, Bulldozer includes support for multiple new CPU instructions, but AMD’s ability to convince developers to adopt them and recompile code for optimum performance is limited.
Click to expand...

Are you just blind or do you need stuff to be posted in BOLD 2 times before you notice it ?
 
Pieter3dnow said:
You have some balls posting this while this is from the article itself

Are you just blind or do you need stuff to be posted in BOLD 2 times before you notice it ?
Click to expand...


yes where are they used? answer the question......
 
a3venom said:
was Bulldozer the worst failure ever in the PC industry? or have there been worse ?
Click to expand...

I would say that Intel's Netburst and Apple fooling people into believing that their mediocre hardware and software is somehow worth the price premium tops the list of PC industry fails.
 
DejaWiz said:
I would say that Intel's Netburst and Apple fooling people into believing that their mediocre hardware and software is somehow worth the price premium tops the list of PC industry fails.
Click to expand...

Apple is probably the biggest WIN in the history of mankind. Capitalizing on the stupidity of the masses, companies can learn from Apple 100 years from now.
I really hate them but they are the biggest geniuses ever.
 
a3venom said:
Apple is probably the biggest WIN in the history of mankind. Capitalizing on the stupidity of the masses, companies can learn from Apple 100 years from now.
I really hate them but they are the biggest geniuses ever.
Click to expand...

Wrong kind of genius though.

We admire geniuses who create new things that are leaps in technology.

People who come up with new genius ways to fool people are pretty much scum.
 
Last edited:
a3venom said:
was Bulldozer the worst failure ever in the PC industry? or have there been worse ?
Click to expand...
Bulldozer is a runaway success compared to some of the gargantuan failures in the industry. Take Windows Vista for example, which supposedly cost Microsoft $6e9 yet never even reached 20% market share. Or how about the Amiga? Far in advance of any other PC in the 80's, it was poised to dominate the industry in the 90's, but then... nothing.
 
Zarathustra[H];1041919054 said:
Wrong kind of genius though.

We admire geniuses who create new things that are leaos in technology.

People who come up with new genius ways to fool people are pretty much scum.
Click to expand...

Still, a genius.
I hate them, You hate them, but if we had such an opportunity to legally cash in and make Billions$, you would take it without blinking.

Mr Evil said:
Bulldozer is a runaway success compared to some of the gargantuan failures in the industry. Take Windows Vista for example, which supposedly cost Microsoft $6e9 yet never even reached 20% market share. Or how about the Amiga? Far in advance of any other PC in the 80's, it was poised to dominate the industry in the 90's, but then... nothing.
Click to expand...

I meant in the CPU world, sorry. I don't think anything was worse than Vista and my first PC was Windows ME when i was 7
 
a3venom said:
I meant in the CPU world, sorry. I don't think anything was worse than Vista and my first PC was Windows ME when i was 7
Click to expand...

Vista wasn't all that bad on new builds, like ME, and just like ME, upgrading to Vista was very painful.

I didn't upgrade to 7 until close to SP1.
 
a3venom said:
...and my first PC was Windows ME when i was 7
Click to expand...

Gee, thanks for making me feel old :p

I was starting my Sophomore year in college when ME was launched :p

The first PC's I used were my parents XT and 386 (respectively) monochrome Toshiba based laptops with DOS 3.1 (if memory serves) in the late 80's.

My first PC of my own was my desktop 12mhz 286, with a 13" VGA monitor, 256kb VGA graphics, 4MB of RAM, a 5.25" floppy drive and a 20MB hard drive in 1991 when I was 11 :p
 
Last edited:
Zarathustra[H];1041919223 said:
Gee, thanks for making me feel old :p

I was starting my Sophomore year in college when ME was launched :p
Click to expand...


LOL pretty much the same here.
 
I think I'm halfway in between you guys age-wise, but feel the same, haha.

First computer at home was a screamer of a 486 DX/33 in 3rd grade. Got used to making boot disks for DOS to run games in HIMEM. (I think?)
 
I had a lot of different computers as a kid, started p3, p4 then AMDs
I knew a lot about it for my age as a kid, but i built the first one of my own after college, as a switch from xbox 360.
It was really a fun experience, glad i never got a PS4.
Just made it so i can play all the single player titles (piratebay) , FIFA and DOTA / WoW
 
DejaWiz said:
I would say that Intel's Netburst and Apple fooling people into believing that their mediocre hardware and software is somehow worth the price premium tops the list of PC industry fails.
Click to expand...
actually NetBurst wasn't that bad, especially Northwood with HyperThreading. Due to ridiculously high clock P4 could stand its ground and thanks to HT system responsiveness was on entirely different level. Even though almost all applications were single threaded it still meant that application running at 100% cpu time was only hogging one processor thread and second one was free to use by both programs and devices/drivers so no irritating choppiness which was plaguing all single core processors.

Try to run modern software such as browse net on Athlon XP 3200+ and do the same on eg. Pentium 4 HT 3GHz. Difference in will be staggering, like comparing processors from different eras!

Bulldozer is much bigger fail than NetBurst ever was. If it had at least the same difference in clock speeds as Pentium 4 had compared to Athlon K7 then its clocks would be in 5.5GHz to almost 7GHz range. That would make it win lots of benchmarks and make it actually look like a decent processor.
 
Mr Evil said:
Bulldozer is a runaway success compared to some of the gargantuan failures in the industry. Take Windows Vista for example, which supposedly cost Microsoft $6e9 yet never even reached 20% market share. Or how about the Amiga? Far in advance of any other PC in the 80's, it was poised to dominate the industry in the 90's, but then... nothing.
Click to expand...
Vista did dominate market because Windows 7 *IS* Vista.
If you did actual comparison, best on two identical computers running next to each other you would see that Windows 7 is Vista with different theme and some other visual changes and few features removed.

Even Windows 8 and Windows 10 are still based on the same underlying technology.

If Windows 7 was released instead of Vista then it would fail too because main issue was lack of drivers, software incompatibility and higher resource usage than XP. Windows 7 never fixed any of those Vista issues, they fixed themselves and Win7 just came later to the scene when drivers were already available, software updated and typical hardware (cheap laptop) was able to run it without issues.

What do you suppose was cost of making Windows 7 out of Vista? I will answer you: whatever cost there was it was spent mostly on advertising. No benchmark showed any significant improvement of Win7 compared to Vista SP1 and SP2, some even showed decrease in performance. But people badly wanted to believe Win7 is more like XP and they did despite it being the same Vista they bashed a while ago.
 
NaroonGTX said:
I dunno how no one mentioned Itanium yet lol.

Also LOL @ saying Bulldozer was worse than Netburst.
Click to expand...


Itanium itself was not a bad product comparing it to the x86/x64 chips out at the time, it had the performance to stay up with the best of em but it was a software problem, where no software was built for it, and to get the marginal performance increase from Itanium if they reprogrammed current software for it was not viable because of the cost of the rewrite. This is why MS opted to support AMD x64 over Intel's itanium and this is what caused it to fail.
 
That's some nice revisionist history. Itanium crashed and burn two years before AMD even had its first x64 processors out in market. Performance of the initial Itaniums were mediocre compared to pre-existing RISC and CISC processors that were available. It was most certainly both a hardware AND software problem. It had to emulate x86 instructions which was bad enough, and no one wanted to support it due to its lackluster performance overall. The thing also suffered from numerous delays. The thing was a laughing stock before it even launched. The arrival of AMD's x86-64 ISA definitely sped things along, but wasn't the sole catalyst by a long shot.
 
NaroonGTX said:
That's some nice revisionist history. Itanium crashed and burn two years before AMD even had its first x64 processors out in market. Performance of the initial Itaniums were mediocre compared to pre-existing RISC and CISC processors that were available. It was most certainly both a hardware AND software problem. It had to emulate x86 instructions which was bad enough, and no one wanted to support it due to its lackluster performance overall. The thing also suffered from numerous delays. The thing was a laughing stock before it even launched. The arrival of AMD's x86-64 ISA definitely sped things along, but wasn't the sole catalyst by a long shot.
Click to expand...

http://www.eejournal.com/archives/articles/20130306-itanium/

yes that is what I kinda ment rewriting the software for the small amount of performance increase was not viable. Without the software rewrite it wouldn't get the performance to meet up with what its potential was.
 
JMccovery said:
Vista wasn't all that bad on new builds, like ME, and just like ME, upgrading to Vista was very painful.

I didn't upgrade to 7 until close to SP1.
Click to expand...

This. People that had problems with Vista tried to run it on older hardware. I had no issues with Vista. If you had a Core 2 or Athlon 64 X2 system with 2GB or more of RAM and a modern GPU it was fine.
 
variant said:
Was this posted? I know it's been known for at least a couple days now.

AMD Zen And K12 Designs Completed And Taped Out
Click to expand...

Already confirmed as false, but wfctech hasn't updated/pulled the source article yet (of course).

razor1 said:
http://www.eejournal.com/archives/articles/20130306-itanium/

yes that is what I kinda ment rewriting the software for the small amount of performance increase was not viable. Without the software rewrite it wouldn't get the performance to meet up with what its potential was.
Click to expand...

Itanium was meant to eventually replace X86, and the X86 mode was strictly for legacy applications. The root problem was POWER/SPARC were better chips for it's initial market, and then AMD killed it in desktops via X86-64. As an architecture, Itanium is simply better then X86-64, as it did away with a lot of X86's legacy problem we're still stuck with today.

Mr. Bluntman said:
This. People that had problems with Vista tried to run it on older hardware. I had no issues with Vista. If you had a Core 2 or Athlon 64 X2 system with 2GB or more of RAM and a modern GPU it was fine.
Click to expand...

Agree. I never had any issues with Vista, but then again, an OCd QX9650 was top tier at the time. If you had the necessary HW to run it well, it was rock solid compared to XP.
 
XoR said:
Vista did dominate market because Windows 7 *IS* Vista.
If you did actual comparison, best on two identical computers running next to each other you would see that Windows 7 is Vista with different theme and some other visual changes and few features removed.

Even Windows 8 and Windows 10 are still based on the same underlying technology.

If Windows 7 was released instead of Vista then it would fail too because main issue was lack of drivers, software incompatibility and higher resource usage than XP. Windows 7 never fixed any of those Vista issues, they fixed themselves and Win7 just came later to the scene when drivers were already available, software updated and typical hardware (cheap laptop) was able to run it without issues.

What do you suppose was cost of making Windows 7 out of Vista? I will answer you: whatever cost there was it was spent mostly on advertising. No benchmark showed any significant improvement of Win7 compared to Vista SP1 and SP2, some even showed decrease in performance. But people badly wanted to believe Win7 is more like XP and they did despite it being the same Vista they bashed a while ago.
Click to expand...
If Windows 7 is Vista, then ME is 98 and 2000 is XP. The differences may be small or cosmetic, but a lot of small differences add up to a huge pile of crap. Maybe it's been so long that people no longer remember what put people off Vista, like the hypersensitive UAC that wouldn't even let you delete a start menu shortcut, or how trying to copy files slowed everything down to a halt. Yes, some of it was due to users' inadequate hardware (XP faced the same problem on release, and MS had to release two SPs before it was unequivocably better than 98), but Windows 7 used substantially less RAM and disc space, making it a real improvement, especially for lower-end systems.
 
gamerk2 said:
Already confirmed as false, but wfctech hasn't updated/pulled the source article yet (of course).



Itanium was meant to eventually replace X86, and the X86 mode was strictly for legacy applications. The root problem was POWER/SPARC were better chips for it's initial market, and then AMD killed it in desktops via X86-64. As an architecture, Itanium is simply better then X86-64, as it did away with a lot of X86's legacy problem we're still stuck with today.



Agree. I never had any issues with Vista, but then again, an OCd QX9650 was top tier at the time. If you had the necessary HW to run it well, it was rock solid compared to XP.
Click to expand...

CEO herself said that many of the next gen FinFet processor designs were finished. So I have no reason to believe that its not taped out. I believe the work on the processor was done when Keller left. I am sure they already have samples in the labs. If the claims are true of performance AMD wont hold back samples, I can bet Anandtech will be doing an early sample review if performance is on par with amd claims. They will create the hype.
 
Pieter3dnow said:
Where was it confirmed false ?

This is the source article , normally he keeps track of this

http://dresdenboy.blogspot.de/2015/10/amd-zen-and-k12-arm-tapeouts-confirmed.html
Click to expand...

Yeah, and it's pretty unambiguous when the CEO of AMD says:

So as we stated in the financial analyst day, we had a target of 40% IPC performance of Zen over our previous generation. We believe we are on track for that.
Relative to process technology, we have taped out multiple products to multiple fabs in FinFET and we believe that they’re also on track in terms of overall ramp. So we continue to focus on both of those aspects, both the architecture and the process technology, but so far, so good.​
 
Already confirmed as false, but wfctech hasn't updated/pulled the source article yet (of course).
Click to expand...

Confirmed where? Nowhere. Zen taped out already quite a while ago. In fact it won't be long before early sample "leaks" start appearing.
 
There has been really no confirmation from AMD outside of generalities that Zen has been taped out, but to get Zen out by end of next year they would have to tape out around now. Validation takes time quite a bit longer than for GPU's.
 
new carrizo sku to replace the embedded kaveri (bald eagle).
listed as having a DDR4 EEC memory controller as well as DDR3.

new spin of the carrizo die, or an enabled feature of the existing die?
 
variant said:
Yeah, and it's pretty unambiguous when the CEO of AMD says:

So as we stated in the financial analyst day, we had a target of 40% IPC performance of Zen over our previous generation. We believe we are on track for that.
Relative to process technology, we have taped out multiple products to multiple fabs in FinFET and we believe that they’re also on track in terms of overall ramp. So we continue to focus on both of those aspects, both the architecture and the process technology, but so far, so good.​
Click to expand...


And I'm pretty sure that if given the time I can dig up a few posts by JF-AMD stating that Bulldozer would certainly be much faster than Phenom II. :p

I am cautiously optimistic, but I have been burned by AMD's pre-launch statements before, so I am reserving any comments until we have actual figures from actual silicon tested by third parties in hand.
 
Zarathustra[H];1041923444 said:
And I'm pretty sure that if given the time I can dig up a few posts by JF-AMD stating that Bulldozer would certainly be much faster than Phenom II. :p

I am cautiously optimistic, but I have been burned by AMD's pre-launch statements before, so I am reserving any comments until we have actual figures from actual silicon tested by third parties in hand.
Click to expand...

I described what happened in earlier post he fessed up about it on SA forums (if my memory recalls it correctly) it was a funny read ;) .
Came down to him not understanding what the engineers were talking about ;) .

But that statement of him saying that about Bulldozer was contained to forums, never saw it posted on front page news...
 
Zarathustra[H];1041923444 said:
And I'm pretty sure that if given the time I can dig up a few posts by JF-AMD stating that Bulldozer would certainly be much faster than Phenom II. :p

I am cautiously optimistic, but I have been burned by AMD's pre-launch statements before, so I am reserving any comments until we have actual figures from actual silicon tested by third parties in hand.
Click to expand...

You are comparing some random AMD employee's opinion on a CPU posted on a forum to a CEO informing their investors?
 
variant said:
You are comparing some random AMD employee's opinion on a CPU posted on a forum to a CEO informing their investors?
Click to expand...

JF-AMD is John Fruehe. He was a director of product marketing at AMD.

Sure, I know what you mean, investor disclosures are sacred and all that, but its not like he was just some Joe shmoe leaker who claimed he worked for AMD and leaked shit on forums.

He was pretty much the public face of AMD for the year proceeding the bulldozer launch.
 
Last edited:
Zarathustra[H];1041924340 said:
JF-AMD is John Fruehe. He was a director of prodict was a marketing at AMD.

Sure, I know what you mean, investor disclosures are sacred and all that, but its not like he was just some Joe shmoe leaker who claimed he worked for AMD and leaked shit on forums.

He was pretty much the public face of AMD for the year proceeding the bulldozer launch.
Click to expand...

So he was PR.
 
Out of relevance, Dirk Meyer (CEO) July 2010 (Bulldozer consumer release Oct 2011) -
In the second quarter this year we also taped out the first 32 nm product based on our new high performance Bulldozer CPU core. We plan to begin sampling our Bulldozer based server and desktop processors in the second half of this year and remain on track for 2011 launches. These new processors will deliver significant performance improvements to the AMD platform.
Click to expand...
 
limitedaccess said:
Out of relevance, Dirk Meyer (CEO) July 2010 (Bulldozer consumer release Oct 2011) -
Click to expand...

To be fair, Bulldozer did offer significant increases...











...in mostly INT workloads spread across all 8 threads. A highly cherry picked scenario.
 
You must log in or register to reply here.
Back
Top