I have recently read in a maximum pc magazine that the new amd phenom processor has bombed pretty hard. when compared and benchmarked the only thing that it performed good in was a simple decoding bench, and this wasn't compared against the most high end intel processors it was compared against a intel qx6700 and a intel q6600 some of the first intel quadcore processors. Just wondering if anyone has a phenom processors and maybe has a different tale to tell then maximum pc? Seeing that intel is going to be releasing there 45nm processors very soon its starting to seem like amd isn't going to get out of this rut anytime soon. If you would like you can find the full articles in the February issue of maximum pc!

No, your pretty much right on that one. You can go look up the reviews on the Phenom Proc and they all do say the same thing

Well yes, compared to Intel's quads, they suck. But the cheapest Phenom still goes for about $190, and most expensive about $240, and Intels cheapest C2Q is about $255. For 4 cores, I think its pretty nice. Would I get it myself? Nope, Intel fanboy

From what I've read so far, AMD is on an "Eco Trip" making their new line more energy efficient at a cost to performance. From the reviews I've read, the reviewer notes that there is power savings but then dwell on the performance issues. To me, not many people are concerned with power savings as they are with CPU performance.

Acctually at FIRST amd is bombing this. But I like to call it the calm before the storm. The reason the chips suck at first is because they are bottle necked at stock speeds. All the latencies in the chip are too low for the chip to even function efficiently on its own.
However, once the 2.6ghz barrier has been breached, alot of users are reporting back with an almost impossible 100%+ increase in performance per clock.

the title to this thread on another forum speaks volumes

"Phenom offers more than 100% clockscaling?!"

and with excerts like

"another interesting review was about multicore gaming performance. i think the phenom @ 2.4 did better than the qx9770 stock in ut3, which is heavily multithreaded. too tired to link, head over to lostcircuits review"

"

Originally Posted by KTE Guys, check this out in an official review (trustworthy). I spotted something odd in CB10 review by Xbit-Labs. I've seen many oddities in a few reviews around, but not the time to spend on them nor discussing them but I'll point this out briefly. All NB, RAM, HT frequencies remain the same below, only the CPU speed/multi changed for higher speeds. Thus, pure CPU MHz scaling theoretically (attached image). From 2.2GHz to 2.6GHz, 400MHz increase there is a 1087 CB difference. From 2.6GHz to 3.0GHz, 400MHz increase, there is a 1480 CB difference. Is this what Gary of AnandTech and the rest of the reviewers talked about with needing plus 2.8GHz Phenom to do well? To put it clearer; From 2.2->2.6GHz, there's a +18.18% clock change. And from 7114->8201 CB, there's a +15.28% performance change. BUT From 2.6GHz->3.0GHz, there's a +15.38% clock change. And from 8201->9681, there's a +18.05% performance change. Which is obviously above what is usually possible, i.e., above 100% clock scaling. Don't know which numbers I can trust here, but I'll see what I can get with my BE and if those numbers are correct. I can verify the 9500, 9600, 9700, 9900 numbers are correct though from my own runs (although I had higher NB speed/HT).

It stands that the phenom processors will do VERY well, once AMD figures out how to get the clock speeds up on these things.

Also its worthy to note that the infamouse "TLB" Bug that was bane to amd apon release has yet to be seen or verified in a normal user environment.

If you have the chance read up on the article from Februaries maximum pc it talks a lot about the tlb bug and what it actually does the main reason why they dont preform very well is that there is a bios fix for the bug but it cuts cpu power anywhere from 0-60% what the bug does it there is a random chance that when your useing your computer it locks up the bug i believe is located somewhere in the cpu cache the reason why the 2 higher clocked phenoms havnt released yet is because there trying to fix the bug at the silicone level if you dont install the fix to your bios you will have better performance gains than most are posting but you will have more stability issues. it just makes me angry that amd takes all this time to make a true quadcore processor and its horrible at the get go the only thing that i can see so far going good for it is the release of the 790fx chipsets and the ability to use 4 radeon hd 3870's in crossfire. me personally im a amd fanboy but im not too optimistic about this new line of processors who knows maybe they will come out with a high end line like the fx's were for the ath64's that will address all of these issues?

Quote: Originally Posted by xgeko2 If you have the chance read up on the article from Februaries maximum pc it talks a lot about the tlb bug and what it actually does the main reason why they dont preform very well is that there is a bios fix for the bug but it cuts cpu power anywhere from 0-60% what the bug does it there is a random chance that when your useing your computer it locks up the bug i believe is located somewhere in the cpu cache the reason why the 2 higher clocked phenoms havnt released yet is because there trying to fix the bug at the silicone level if you dont install the fix to your bios you will have better performance gains than most are posting but you will have more stability issues. it just makes me angry that amd takes all this time to make a true quadcore processor and its horrible at the get go the only thing that i can see so far going good for it is the release of the 790fx chipsets and the ability to use 4 radeon hd 3870's in crossfire. me personally im a amd fanboy but im not too optimistic about this new line of processors who knows maybe they will come out with a high end line like the fx's were for the ath64's that will address all of these issues?

I really have to disagree with you on a few points here.
The bios fix only hinders up to a 10% hit in performance.

Also the TLB bug doesnt matter to the normal user one bit, as it only occurs in a SUPER rare condition in the Shared L3 cache under heavy virtualization use of the processor with nested memory paging and HIGH utilization for long periods of time. IE SERVER USE.

ok first off straight out of the review from maximum pc which is non-biased, and quoted from AMD the TLB(translation lookaside buffer) bug exists in a small cache used by the cpu to manage memory and the L3 cache. AMD says under heavy work loads such as virtualization the TLB bug can cause the system to Hard Lock which is the reason why they pulled the 2.4ghz and 2.6ghz chips. Also quoted from AMD initial statements gave the impression that the erratum affected only virtualization, which is a server-class application and an uncommon use for a desktop CPU. In truth, the erratum can cause instability with desktop-style usage patterns. also a full article on the performance drop is listed here which is much higher then a 10% hit with the tlb patch Phenom TLB patch benchmarked - The Tech Report - Page 1 the average is about 19% over all the synthetic and realworld benchmarks but if you look at just some of the real world benchmarks things like just useing a web browser your looking at a 57% performance hit! now i love amd processors but they better do something soon i was really looking forward to this chips release.*edit* Also if u look at the memory performance hit its taking a 37% performance hit with the patch and is also taking a 50% hit in memory latency performance.

Quote: Originally Posted by Highperf15 From what I've read so far, AMD is on an "Eco Trip" making their new line more energy efficient at a cost to performance. From the reviews I've read, the reviewer notes that there is power savings but then dwell on the performance issues. To me, not many people are concerned with power savings as they are with CPU performance.

But think about it. The whole world is now going on a Eco Trip, From Fuel efficient cars to using solar power or other renewable energy sources. This is really a Untapped Market for CPU procs. The pay off may take a while, but it may be AMD's ace in the hole.

But besides that, Maybe users will start to consider power usage. Less Power Usage means A Smaller PSU and more money to spend on other parts.

Perhaps AMD is "bombing it" at the high end, and although many users will look at the high end performance benchmarks even though the vast majority of chips sold will be far lesser in power, they'll still choose the company with the best racing program, not necessarily the best platform for the money.

Call it "eco-friendly" call it "cost-efficient", it still comes down to bang for the buck, and unless you actually need the power of a top tier CPU, AMD still produces great chips for the average user, or for corporations that truly can benefit from 1000+ machines all using less electricity. Having never once purchased the top processor available at any given time, count me as an average user very content with their products, especially at the price range that fits my meager budget, for both price and power usage. That's not about an "Eco-Trip", that's about spending my money wisely instead of being so brand loyal that I am compelled to spend more than I really need.

BTW, I have both Intel and AMD systems in my network, and they all went under the same scrutiny when hardware was chosen.

Oh, and xgeko2, if you think Max PC is non-biased...I have some RAMBUS I can make a sweet deal on just for you.

Quote: Originally Posted by Wordbiker Perhaps AMD is "bombing it" at the high end, and although many users will look at the high end performance benchmarks even though the vast majority of chips sold will be far lesser in power, they'll still choose the company with the best racing program, not necessarily the best platform for the money. Call it "eco-friendly" call it "cost-efficient", it still comes down to bang for the buck, and unless you actually need the power of a top tier CPU, AMD still produces great chips for the average user, or for corporations that truly can benefit from 1000+ machines all using less electricity. Having never once purchased the top processor available at any given time, count me as an average user very content with their products, especially at the price range that fits my meager budget, for both price and power usage. That's not about an "Eco-Trip", that's about spending my money wisely instead of being so brand loyal that I am compelled to spend more than I really need. BTW, I have both Intel and AMD systems in my network, and they all went under the same scrutiny when hardware was chosen. Oh, and xgeko2, if you think Max PC is non-biased...I have some RAMBUS I can make a sweet deal on just for you.

He pretty much summed it up for me.... Oh yeah, can we get some more spaces between our ideas, it hurts my eyes when people don't use spaces.


And stating MAXIMUMPC is unbiased is totally and completely wrong. Every writer/reviewer at any sort of publication has some sort of bias in their 'industry'. Bad differences and things of that nature are what creates it..

idk everytime ive read a issue of maximum pc its been for the most part right on. and i know that the first 2 quadcore phenoms they released are low energy the 9500 and 9600 are at 95 watts but the 9700 is going to be at about 125 watts and the 9900 is going to be at about 140watts still i guess it isnt too bad im just hoping that they get something released that can keep up with the hype the "true" quadcore was sapposed to have.

Part of the "original hype" for AMD's quad was the promise of independent clocking for each core. This is purely a power saving feature, not a max power feature. Personally, I like the idea of not burning extra juice when I don't need it, and I'm sure someone administering a server bank would too. Less juice used = less heat to dissipate and less cost to operate.

Keep in mind that TDP is not calculated the same by Intel and AMD and is not a true comparative measure of actual power usage, temperature tolerance or raw computing power. There are many other factors to consider, and no one test will compare every scenario possible.

News sources are paid to create interest, and often overlook the more mundane yet important details of products or events in favor of what will generate the biggest splash. Just because Max PC tells you what you want to hear...it doesn't make them right. To be fair and open-minded, find other sources...not just those you agree with.

Quote: Originally Posted by Wordbiker Perhaps AMD is "bombing it" at the high end, and although many users will look at the high end performance benchmarks even though the vast majority of chips sold will be far lesser in power, they'll still choose the company with the best racing program, not necessarily the best platform for the money. Call it "eco-friendly" call it "cost-efficient", it still comes down to bang for the buck, and unless you actually need the power of a top tier CPU, AMD still produces great chips for the average user, or for corporations that truly can benefit from 1000+ machines all using less electricity. Having never once purchased the top processor available at any given time, count me as an average user very content with their products, especially at the price range that fits my meager budget, for both price and power usage. That's not about an "Eco-Trip", that's about spending my money wisely instead of being so brand loyal that I am compelled to spend more than I really need. BTW, I have both Intel and AMD systems in my network, and they all went under the same scrutiny when hardware was chosen. Oh, and xgeko2, if you think Max PC is non-biased...I have some RAMBUS I can make a sweet deal on just for you.

Check this out

X-bit labs - AMD Says Customers Demand Low-End Power-Efficient AMD Phenom Chips.

Thats a pretty cool article but i read from amd that the reason why the 9700 and 9900 didnt ship yet is because of going to the b3 steping to fix the tlb errata also im wondering if why its also taking so long to ship is maybe there trying to make a fix for the power consumption because once those chips go over 2.3ghz they start heating up alot which could be a fabrication issue. i honestly could care less if they take another 2-3 months working on the 9700 and 9900 phenom's as long as it address the tlb issue at the silicon issue and the power consumption after they hit 2.3ghz and higher another nice thing is the 9900 is sapposed to have a 4ghz hyper transport =D so it may still shape out good.

don't forget about memory bandwidth.
apparently there was something with Sis Sandra that was reporting incorrect memory benchmarks for Phenoms. from some articles and benchmarks I've seen, phenoms seem to get anywhere from 800MB/s to 2000MB/s more memory bandwidth than an intel qx6850. mind you, these sources seem a little to very sketchy to me, but they could be accurate.

and Im on board with those of you who say maximum pc is biased. but I haven't read one in a year or two.

EDIT: sorry, I didn't notice when this article was from

lol happens personally i think the new amd quad core is a waste of money unlease you were already running on a amd dualcore platform that you can directly upgrade too but thats at this point in time im not saying in the future they wont have a great product because i think with how long they worked on this project they have to have something way better coming out soon i think it may have to do with there so called "spider" platform it would be really nice to have a phenom that isnt buggy i would buy one in a heart beat i really just go with whats on top at the time and so far amd is drudging behind not just in performance but also in power look at the new pynryns from intel 3.oghz dualcore wolfdale is only like 65 thermal watts and the quadcore extreme does come in at about 130 thermal watts, the phenom comming in at 95 watts but from what i hear when you do any overclocking if u can on a phenom the thermal watts skyrocket.

xgeko2, full stops are your friend

Has AMD Bombed, yes... There stuff just doesn't compete with Intel. Are they cheaper than an Intel Quad, yes, but in everyday use a standard C2D will out perform them. Sandra Benchmarks are a joke. Take the cheapest C2D and overclock it and test it against the AMD with real applications. What you'll find is a standard C2D will out perform AMD, regardless of the # of cpu's, and it will out perform the Intel Quad Core also. Until software actually starts utilizing more than 2 cores all of this is just for bragging rights anyways. And when the software does catch up you won't want a AMD cpu. Same thing goes for ATI, right now it seems like they have Great new card. But there way behind in releasing it. The new 9800 GX2 will smoke it in a few weeks. AMD just blows, but what's even worse is they are dragging ATI down with them.

Quote: Originally Posted by Slaymate Take the cheapest C2D and overclock it and test it against the AMD with real applications. What you'll find is a standard C2D will out perform AMD, regardless of the # of cpu's, and it will out perform the Intel Quad Core also. Until software actually starts utilizing more than 2 cores all of this is just for bragging rights anyways.

Quads ftw. Check this out:

AnandTech: Unreal Tournament 3 CPU & High End GPU Analysis: Next-Gen Gaming Explored

I think 20-30 extra FPS is a pretty decent increase considering cores 3 and 4 are only running ancillary helper threads.

But yeah, AMD need to pull the finger out. And if they can't, it will be very sad to see ATi go down with them.