VirtualLarry
No Lifer
- Aug 25, 2001
- 56,587
- 10,227
- 126
Well, I am just waiting for Windows to run on this and I will be happy
http://en.wikipedia.org/wiki/IBM_zEC12_(microprocessor)
TDP 1800W? Whoa!
TDP 1800W? Whoa!
- Aug 25, 2005
- 3,743
- 28
- 86
That's part of the architecture, isn't it? I don't think IMC improvements are off limits for AMD designers.
guskline
Diamond Member
- Apr 17, 2006
- 5,338
- 476
- 126
sushiwarrior, where did you come up with your percentage figures? They seem high. I'm going to paste a partial quote from TechSpot's review of the 8350 as an example of what I mean
"Unfortunately, that appears to be true. Besides WinRAR and Fritz Chess, the FX-8350 was at most 6% faster than the FX-8150 in all of our tests, despite being clocked 11% faster (though we should keep in mind that both chips have a max turbo frequency of 4.2GHz). Similar results were seen when comparing the FX-6300 and FX-6100, though the margins were even slimmer here as the Piledriver chip is only clocked 6% higher.
Honestly, despite what we saw with Trinity, we still expected more of the new FX series. Unfortunately, we only had two days to test these new processors, which obviously isn't a lot of time, and much of it was spent trying to determine if there was an issue with our test bed. Nonetheless, we logged consistent results and have no reason to doubt them, despite AMD's press material claiming up to 23% more performance."
Were you relying on AMD's press releases to substantiate your percentages? I think they are a tad high.
I'm hardly an expert, but I owned a 8150 and now own a 8350 (also 8320) and though they are improved, I'm confused with a 20% claim of IPC improvement.
It's ~300W per chip, the 1800W comes from the performance of the cooling solution (I believe the cooling solution covers multiple chips). Just for clarification.
I think that every company has to go through a rough times, both Intel and AMD were there several times.
It's getting annoying that all the fanboys of either company still claim their products are somehow better. Right now yes, the desktop Intel offerings are much better within power consumption, IPC and performance for high end market. However AMD has good offerings with their mobile APUs right now. I also find AMD APUs to be better choice for mainstream computers because you don't need that fast CPU, but much powerful GPU to have under the hood is definitely useful for occasional gaming or multimedia. They recently won contracts to supply chips for all major consoles and they are proceeding to gain more contracts for mobile OEMs.
The BD was a big fail, which everyone knows, but so Intel did the same grade fail if not even worse with the disastrous Prescott and Northwood P4s, but since the debut of conroe, seems like everyone forgot about Intel's mistakes way too fast.
AMD also unlike Intel despite the situation it is in, didn't cut on the product quality like Intel did with replacing soldering with TIM along with bad spacing between the die and IHS and also crappy boxed coolers. On my last AMD rig, I used stock cooler for hardcore gaming for the first 2 years I had it, it was silent, flawless operation with no overheating issues. The intel ones are thrown out even with many users who don't OC or have locked CPUs because of the noise they are generating and low cooling output.
The situation on the market also changed very much lately to the point that face to face competition is(as many people are claiming AMD is not capable to), is not possible nor relevant anymore. The general performance of the computers in the past was a major issue and drove both companies to compete on IPC-scale most of the time ignoring the power consumption and heat generation which was until very lately just increasing and increasing, to record heights. The faster newer CPUs from competition meant that your system booted faster, the games were running better and it was overall more responsive and allowed the basics tasks to be done more pleasantly and faster. This is long gone, optimizing power consumption and getting increase in specific sets of applications today is much more complex and is making companies to directly not compete with each other.
I would say, better is the exact CPU for exact use not company making it.
It's getting annoying that all the fanboys of either company still claim their products are somehow better. Right now yes, the desktop Intel offerings are much better within power consumption, IPC and performance for high end market. However AMD has good offerings with their mobile APUs right now. I also find AMD APUs to be better choice for mainstream computers because you don't need that fast CPU, but much powerful GPU to have under the hood is definitely useful for occasional gaming or multimedia. They recently won contracts to supply chips for all major consoles and they are proceeding to gain more contracts for mobile OEMs.
The BD was a big fail, which everyone knows, but so Intel did the same grade fail if not even worse with the disastrous Prescott and Northwood P4s, but since the debut of conroe, seems like everyone forgot about Intel's mistakes way too fast.
AMD also unlike Intel despite the situation it is in, didn't cut on the product quality like Intel did with replacing soldering with TIM along with bad spacing between the die and IHS and also crappy boxed coolers. On my last AMD rig, I used stock cooler for hardcore gaming for the first 2 years I had it, it was silent, flawless operation with no overheating issues. The intel ones are thrown out even with many users who don't OC or have locked CPUs because of the noise they are generating and low cooling output.
The situation on the market also changed very much lately to the point that face to face competition is(as many people are claiming AMD is not capable to), is not possible nor relevant anymore. The general performance of the computers in the past was a major issue and drove both companies to compete on IPC-scale most of the time ignoring the power consumption and heat generation which was until very lately just increasing and increasing, to record heights. The faster newer CPUs from competition meant that your system booted faster, the games were running better and it was overall more responsive and allowed the basics tasks to be done more pleasantly and faster. This is long gone, optimizing power consumption and getting increase in specific sets of applications today is much more complex and is making companies to directly not compete with each other.
I would say, better is the exact CPU for exact use not company making it.
Last edited:
But it'll be hard to duplicate the one-time gain from going off-chip to integrated and shaving several tens of ns of latency.
As already linked in this thread: Clearly more than "6%" increases. Where is that text from? They mention it is tested with limited time, and they had issues with their test bed. Sounds like a bit of questionable source.
I think AMD is easily capable of the 8150->8350 improvements over again, which would lead to a much more competitive processor (and also hopefully a more power efficient one, with the half node shrink).
Forget techspot and rely on serious sites that show
actual numbers on a clock for clock basis....
http://www.hardware.fr/articles/880-6/bulldozer-vs-piledriver-4-ghz.html
On second thoughts , how much from K6 to K7 ?
Or from Athlon to Athlon XP , and from this one
to Athlon64 and then to K10.?..
Truth is that each of thoses evolutions brought
big IPC gains.
Last edited:
- Aug 25, 2005
- 3,743
- 28
- 86
Well, I'm not an expert at all but from what I've read and heard from more knowledgeable people there is quite a bit of improvement available in AMD's IMC and Cache structure before they hit theoretical limits. Intel's structures in those areas are certainly superior then AMD's according to just about everyone which would confirm the potential available.
If you take out games, the improvement is 7-8 % IPC on average. With games it's about 9-10% on average. This is about what I remembered it to be, although I think with a broader range of games it would be higher (most games showed 15-20% improvement). Estimating 20% was certainly way too high, though the improvement given the very short amount of time and same process was still quite impressive.
Steamroller is on another level in respect to tweaks than Piledriver
was re BD , so it should indeed bring more increasement than PD ,
we ll know more in a few months , there will be forcibly some
leaks since there s already ES in the wild , for the time
i keep on 20% MT and 15% ST figures.
was re BD , so it should indeed bring more increasement than PD ,
we ll know more in a few months , there will be forcibly some
leaks since there s already ES in the wild , for the time
i keep on 20% MT and 15% ST figures.
Montosaurous
Member
- Feb 18, 2013
- 30
- 0
- 0
Well your answer is right there. 3.6 ghz vs 4.0 ghz. Performance increased quite a bit but so did clockspeed (a lot to do with manufacturing issues with BD). IPC increase wasn't nearly 20% (clockspeed increased about 10%).
And efficiency is just barely better than thuban with 8350.
Ventanni
Golden Member
- Jul 25, 2011
- 1,432
- 142
- 106
I don't speak as any kind of "fanboi", but I do think a substantial increase in IPC is quite possible from the BD architecture. What I don't think, however, is that AMD is ever capable at doing it at the same power draw. So basically what I'm saying is, we might see Sandy Bridge level IPC out of Excavator (which would be an impressive feat considering it has a long ways to go to match Intel's IPC advantage), but it'll do it at a much higher power draw.
dlamb2471
Member
- Dec 21, 2010
- 56
- 0
- 66
I imagine we'll get some increase, but I'm prepared to be disappointed. I've been an AMD fan since the first Athlon CPU's came out, but Sandy Bridge finally convinced me to go back to Intel. We all benefit from actual competition...imagine what Intel is holding back based on AMD's lack of performance. Same thing goes for Nvidia vs AMD/ATI graphics...let us all hope for continued strong competition.
That being said...I expect very little.
That being said...I expect very little.
- Jan 31, 2000
- 5,831
- 1,043
- 126
I don't expect a miracle, but i am definitely more excited about Kaveri and Steamroller chips than any other AMD cpu release.
Looks like a step in the right direction anyways.
If it's not what i hope it is, then it's back to Intel as i'm more than happy with my AMD 960T for now.
Looks like a step in the right direction anyways.
If it's not what i hope it is, then it's back to Intel as i'm more than happy with my AMD 960T for now.
- Aug 25, 2005
- 3,743
- 28
- 86
That's relevant because? As far as actual performance it's Instructions Per Second that matters.
dlamb2471
Member
- Dec 21, 2010
- 56
- 0
- 66
The big question remains...does AMD have the resources to compete? I hope like hell they do, but I suspect that they don't. The rise of ARM cpu's worries me that we may well soon end up in a world (we're already there aren't we?) where Intel is the only provider of workstation class CPU's. High end performance may be dying off in popularity - Windows and Linux have run well enough for years on the hardware we have. But creation efficiency will always be boosted by performance increases until the time where any computing task is nearly instant. It could be graphics rendering, compiling, video encoding...you name it. When you're creating something and you have to wait on your computer, that is potential downtime. If I'm compiling a project and I need to test a change that I've made, let's say that I made a change that requires a complete rebuild. At that point in time, I literally have to wait on the machine, because I'm ready to test the changes that I have made.
Sorry for getting the thread off track with my musings...let's say...10% boost is what I expect from this next revision from AMD.
Sorry for getting the thread off track with my musings...let's say...10% boost is what I expect from this next revision from AMD.
That lovely efficiency graph you posted proves my point exactly
8150->8350 showed gains in efficiency and speed. No reason that steamroller won't also show gains. It's just how much that matters.
That lovely efficiency graph you posted proves my point exactly
Do you not see where they took a step back from thuban?
The gain from thuban -> 8350 is half of the gain from 8150 -> 8350.
AMD was simply making up lost performance.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 24K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter