![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Things AMD can do to take on Conroe...
- Thread starter freeloader1969
- Start date
MrGuvernment
Fully [H]
- Joined
- Aug 3, 2004
- Messages
- 21,919
wrangler said:
I dont think that alone would help AMD take out conroe.
Merom has talks of 4mb cache as well: anand:
More for intel, but i think this migjht only be on the mobile line and this is i think 2007 tech:
Now if AMD has something like this for their desktop chips i could see a serious boost in performance.....
I'd say keep with their cache scheme, it is very fine tuned, where Intel shares the same copy of instructions on each level AMD dosn't and the 128KB+xMB work together better, combined with the short pipeline and better local branch prediction. I don't see a reason why, adding more cache increases the latency. I think AMD should put a focus on having the abillity of opening a long stage pipeline for media and work to create a specialized instruction set meant for processing ingame physics and AI.freeloader1969 said:What does everyone think AMD will have to do to be competitive against Conroe? Some suggestions would be great.
Here's what I was thinking. Take the K8L (with extra floating point units), add a 4mb L2 cache (2mb per core), possibly some L3 cache, and if it's even possible to do in a few months, add 2 more stages to the Athlon 64 core.
I'm in no way a CPU engineer, so go easy on me!
My logic was that the extra stages may help the existing arcitecture scale over 3ghz. I realize the L3 cache is wishful thinking due to 90nm fabrication. Maybe at 65nm we'll see some. I also don't believe that the switch to DDR2 is going to make that much of a difference in current performance. A64's have never been bandwidth hungry anyhow.
Thoughts, suggestions?
stealthy123
2[H]4U
- Joined
- Feb 4, 2005
- Messages
- 3,149
wrangler said:
they licensed zram. It will be a couple years before we see anything from it though.
MrGuvernment
Fully [H]
- Joined
- Aug 3, 2004
- Messages
- 21,919
Obi_Kwiet said:
They already are - well single core ones......
Correct me if I'm wrong on these:
First of all, wasn't SSE developed in order to overcome the P4's lack of a real FPU? I believe I read somewhere that Intel couldn't fit their original P4 design on the die size they used at the time, and so chopped up the FPU.
The only way to overcome the P4 weaknesses was to clocked it up, add more, and longer pipes, which added to other problems, including it's biggest, heat. They also came out w/compilers, specifically written for the P4's, with support for SSEx's, to help overcome it's weaknesses.
Clock for clock, the P3 was always faster than the P4, which is why, I would imagine, among other things, Intel went full circle back to essentially the P3 design.
AMD, on the otherhand, stayed away from the clocking craze and instead went w/a more efficient design, including a full FPU, as well as adding the memory controller, a bigger L2 cache, and x64, onto an essentially Athlon core, when the die size allowed them to do it, w/o sacrificing performance.
Intel having been number one, had to stay ahead, no matter what it took, while AMD simply played catch up. Only this time, I feel, because of Intel's biggest screw up of going w/the P4 design in the first place, AMD passed them. This time, Intel has to, not only play catch up, but get pass AMD again. Also, I though it was stupid of Intel to go full bore w/a their totallly 64 bit chip, rather than going AMD's route of allowing a slow merger by allow both 32 and 64 bit support on their A64, which allows a slow transition of technology, rather than a forced one. Because of this botch up, Intel fell back on that and ended up using AMD's x64 code. You also notice that AMD took their time before going the DDR2 route, since DDR has kept them at pace w/Intel.
I think the biggest problem that both AMD and Intel have to face is the slow pace of the memory industry, playing catch up w/both of them. Due to the much slower memory, they are both forced to have multipliers added to the cpu's in order to compensate for the much slower memory.
First of all, wasn't SSE developed in order to overcome the P4's lack of a real FPU? I believe I read somewhere that Intel couldn't fit their original P4 design on the die size they used at the time, and so chopped up the FPU.
The only way to overcome the P4 weaknesses was to clocked it up, add more, and longer pipes, which added to other problems, including it's biggest, heat. They also came out w/compilers, specifically written for the P4's, with support for SSEx's, to help overcome it's weaknesses.
Clock for clock, the P3 was always faster than the P4, which is why, I would imagine, among other things, Intel went full circle back to essentially the P3 design.
AMD, on the otherhand, stayed away from the clocking craze and instead went w/a more efficient design, including a full FPU, as well as adding the memory controller, a bigger L2 cache, and x64, onto an essentially Athlon core, when the die size allowed them to do it, w/o sacrificing performance.
Intel having been number one, had to stay ahead, no matter what it took, while AMD simply played catch up. Only this time, I feel, because of Intel's biggest screw up of going w/the P4 design in the first place, AMD passed them. This time, Intel has to, not only play catch up, but get pass AMD again. Also, I though it was stupid of Intel to go full bore w/a their totallly 64 bit chip, rather than going AMD's route of allowing a slow merger by allow both 32 and 64 bit support on their A64, which allows a slow transition of technology, rather than a forced one. Because of this botch up, Intel fell back on that and ended up using AMD's x64 code. You also notice that AMD took their time before going the DDR2 route, since DDR has kept them at pace w/Intel.
I think the biggest problem that both AMD and Intel have to face is the slow pace of the memory industry, playing catch up w/both of them. Due to the much slower memory, they are both forced to have multipliers added to the cpu's in order to compensate for the much slower memory.
freeloader1969 said:Try and stay on topic guys. I don't think we'll see nano machine guns anytime soon
needmorecarnitine...If you don't have any ideas, then don't contribute. Just skip over this thread. I don't think there's a single CPU engineer on this whole board, atleast no one has ever claimed to be.
Some interesting ideas so far. We have extra cache, stronger IPC, 1 pass vector processing and more FP units. Sounds similar to what Intel has done, minus the FP units.
Hey I'm a CPU Engineer
Well, I probably don't really count since I'm still in grad school!I've read that AMD should be releasing a completely new architecture in 2007. I don't think the current architecture will best Intel's new architecture.
morfinx said:Hey I'm a CPU Engineer
I've read that AMD should be releasing a completely new architecture in 2007. I don't think the current architecture will best Intel's new architecture.
Whatever. Yeah, AMD has been working on a new architecture for the past 1-2 years to make sure its slower than Conroe. josh_1413 said:
Josh, I think you misread my post. I said I think AMD's CURRENT architecture (K7) may not be able to catch up with Intel's new architecture. I was not talking about AMD's (yet to be announced) new architecture, which I'm sure will be a force to be reckoned with.
I'm just as big of an AMD fan-boy as anyone here. In fact one of the big reasons why I chose to go to grad school at my particular school is that I know AMD recruits most of its new engineers from several schools, this being one of them.
USMC2Hard4U
Supreme [H]ardness
- Joined
- Apr 4, 2003
- Messages
- 6,157
The biggest thing they need to do, and which have always lacked....
They need better SIMD performance. They FPU speed is outstanding for classic applications... but not for SIMD stuff (Which everything now is, since the release of SSE)
The Pentium 3 had superior SIMD Performance compared to the Athlon
The Pentium 4 had superior SIMD Performance compared to the Athlon XP and even A64
Now this conroe blows it away...
AMD needs better support and speed for their SSE, SSE2, SSE3 Instruction sets. Since everything is coded for this now... its where all the Conroe's Media performance comes from.
They need better SIMD performance. They FPU speed is outstanding for classic applications... but not for SIMD stuff (Which everything now is, since the release of SSE)
The Pentium 3 had superior SIMD Performance compared to the Athlon
The Pentium 4 had superior SIMD Performance compared to the Athlon XP and even A64
Now this conroe blows it away...
AMD needs better support and speed for their SSE, SSE2, SSE3 Instruction sets. Since everything is coded for this now... its where all the Conroe's Media performance comes from.
SpoogeMonkey
2[H]4U
- Joined
- Feb 20, 2005
- Messages
- 2,934
USMC2Hard4U said:
The way amd implements sse is lacking. Adding sse3 to the venice cpus resulted in ~0% performance increase. I hope it's something as simple as this.
robberbaron
Supreme [H]ardness
- Joined
- Sep 6, 2004
- Messages
- 6,101
josh_1413 said:
It's being made with servers in mind AFAIK. I would imagine if yields on it are good, they might migrate it from Socket F to AM2. But it's gonna contain a rather large die. Only thing I know for sure is Socket F.
DejaWiz
Fully [H]
- Joined
- Apr 15, 2005
- Messages
- 21,826
I heard it would be around the same timeframe as AM2.josh_1413 said:
covertclocker
Limp Gawd
- Joined
- Jun 10, 2004
- Messages
- 498
Mega2 said:
Well, reading a post from Ecplise a little while back, GDDR3 has disgustingly high latencies. Hence the reason why its not used as system ram yet. Quad core is on the horizon too.
[LYL]Homer
Supreme [H]ardness
- Joined
- Jun 7, 2004
- Messages
- 4,209
Given that the A64 does more per clock cycle than the P4 is it possible we'll see a future chip with even fewer mhz, something like a 1ghz chip that could beat Conroe?
morfinx said:Hey I'm a CPU Engineer
I've read that AMD should be releasing a completely new architecture in 2007. I don't think the current architecture will best Intel's new architecture.
you're a cpu engineer?Well enlighten us with how AMD can create a cpu to beat conroe already
The K8 A64 is 3 years old and a 3 year old arc. What do ya think they been doing for the past 3 years? Nothing would be wrong and AM2 Enhanced K8's are not one of them. The new arc is going to be as big as a P4 going to Conroe. K9's will be one hell of a arc for AMD. All they did with the Enhanced K8's is perfect 90nm tech. What intel could only dream of in eff and low power use with 90nm's. K9's arn't only a change from 90nm to 65nm. But a arc change they been working on for as long as 90nm has been around.
They knew 65nm's was coming and they where going to include it into their new arc as soon as they could when IBM found out how to do it. Early AM2's arn't made to impress anybody, they are made for ppl that don't want a cpu that runs more then 125w/175w *Cough* Intel. *Cough* AMD knows they are behind but again what can 90nm do agenst a 65nm arc. 65nm isn't just a process change but it allows for higher clocks with lower heat and lower power use. Also allows a smaller area to put the die on that can perform just as good as a larger die at 90nm's. 65nms is a good amount smaller then a 90nm die and if made just as large can be even faster but not alone it just sets a new bar really cus you can fit more stuff into a smaller area, you can add more of something with out loseing performance. Why intels L2 is getting so damn big now is all the reason of 65nm's, ya could boot with no memory at all if ya wanted to with that much L2. =P J/K
Intel had to change to that 65nm to beat AMD. Still wasn't enough then they thought of conroe a new arc on 65nm's. AMD will do the same thing when moving to 65nm. And it will be even more eff then a intel knowing how AMD is. AMD knows the K8's arn't good enough any more. Now its time to move to K9's they once said when they anounced that new arc on 65nm's. From one AMD fan to another. AMD woot woot woot. Intel fans shouldn't even be looking for posts like this in the 1st place just to get mad at. I mean its the truth ant it. AMD wins the 6 months then intel after that then amd and so on over and over bla bla bla. Besides I'd rather have a eff 67w or 35w X2 cpu then one that runs 100w or 200w like a intel dual core.
They knew 65nm's was coming and they where going to include it into their new arc as soon as they could when IBM found out how to do it. Early AM2's arn't made to impress anybody, they are made for ppl that don't want a cpu that runs more then 125w/175w *Cough* Intel. *Cough* AMD knows they are behind but again what can 90nm do agenst a 65nm arc. 65nm isn't just a process change but it allows for higher clocks with lower heat and lower power use. Also allows a smaller area to put the die on that can perform just as good as a larger die at 90nm's. 65nms is a good amount smaller then a 90nm die and if made just as large can be even faster but not alone it just sets a new bar really cus you can fit more stuff into a smaller area, you can add more of something with out loseing performance. Why intels L2 is getting so damn big now is all the reason of 65nm's, ya could boot with no memory at all if ya wanted to with that much L2. =P J/K
Intel had to change to that 65nm to beat AMD. Still wasn't enough then they thought of conroe a new arc on 65nm's. AMD will do the same thing when moving to 65nm. And it will be even more eff then a intel knowing how AMD is. AMD knows the K8's arn't good enough any more. Now its time to move to K9's they once said when they anounced that new arc on 65nm's. From one AMD fan to another. AMD woot woot woot. Intel fans shouldn't even be looking for posts like this in the 1st place just to get mad at. I mean its the truth ant it. AMD wins the 6 months then intel after that then amd and so on over and over bla bla bla. Besides I'd rather have a eff 67w or 35w X2 cpu then one that runs 100w or 200w like a intel dual core.
freeloader1969
2[H]4U
- Joined
- Dec 13, 2003
- Messages
- 2,238
AMD will counter Conroe with DDR2/K8L. Will it be enough to stay with Conroe or beat it, who knows? IF AMD is able to incorporate a single pass SSE implementation (time is basically their enemy) into the K8L, we'll have a good old fashioned shootout again! In any event, it's good to see Intel back in the game. AMD will have to drop prices in the short term to compete.
covertclocker said:
DDR3 will have similar latencies to GDDR3. Accessing bits in a large DRAM cell is painfully slow, and that's not going to change. The logic surounding accessing the bits can be done very quickly, and reads can be very deeply pipelined. DDR, DDR2, and soon DDR3 take advantage of that by having very fast logic, slow memory cores and just reading more per bits in parallel each clock cycle.
Where PC133 read say 1 bit per cycle, DDR 266 does two reads per cycle (one every half cycle) and DDR2 533 reads 4 bits (2 each cycle from 2 parts of the memory core), DDR3 1066 will read 8 bits per clock cycle for each read operation.
So with DDR3 1006, acessing the memory core is only about as fast as it was in DDR266 (which means at 4 times the clock speed, you have CAS8 or 10 instead of CAS2 or 2.5), but you get 4 times the bandwidth of DDR266 without having to speed up the actual reads by 400%. (And actually the total latency of a read is a bit less because all the logic surounding the read happens much faster (in terms of actual time, it takes roughly the same number of clock cycles) than it did with DDR1 ram.
The one in all Thrend. The Conroe debunk thrends is all ya need for your conroe needs. Just wait until real world final product performance issue results come out. That will really make us AMD fan boys laugh at intel for trying to cheat on performance with a shit load of L2 for a large performance gane. No L2 see a FX60 F up a conroe. lol This HYPE will get old real fast.
Serge84 said:
Two small notes, if I'm deciphering this random collection of seemingly Enligh phrases correctly.
1. You can't "cheat" at performance. You can cheat at specific benchmarks by hand-optimizing for them. Adding more cache is no better or worse than any other CPU enhancement. A given enhancement will have certain tradeoffs, or may not do as well in certain circumstances. Calling it "cheating" betrays a fundamental retardation.
2. You keep referring to this conroe debunk thread. It's idiotic, it's like you think it's some sort of canonical refutation of the whole concept of Conroe. Conroe has been winning, by a considerable amount, both synthetic and real world benchmarks of all sorts.
You can claim it's only winning synthetics, and you can claim it's "cheating", but reality would seem to disagree with your assessments.
$BangforThe$
[H]ard|Gawd
- Joined
- Nov 5, 2005
- Messages
- 1,672
Serge84 said:
Nice post very informative . If Conroe can't get the job done I suppose Intel could use this CPU . It beats AMD at everthing.
http://www.anandtech.com/mb/showdoc.aspx?i=2750
Intel has had 2 years of failures.
Itanic.
Pentium IV 4000 recalled. Still haven't put out a working one. Merom delayed till 4th quarter atleast, Intel finally putting out a 64 bit laptop cpu. Lots of paper releases with actual product lagging months behind. Pentium IV extremely expensive and no great improvement editions. Xeon eaten for lunch by Opteron.
The Conroe benchmarks I have seen were very fishy at best. The 'testing' was so controlled that I don't feel the results have any credibility. At best a smoke screen hoping to delay purchases. If the test had any credibility the testers would have been able to look inside the machines, check bios settings etc. Anyone can rig a test to produce just about any results. Let them spend a few days on the machines, look inside, see what drivers are being used etc.
Call me when they do that and then I'll beleave the shit I see. I'm pritty sure that has ya really going there. You would be a fool to beleave benchmarks in such a controlled way of "testing" things. It makes me sick how much of a lie conroe is. And ppl can be suckered enough to beleave how good it is with such questionable credibility.
Itanic.
Pentium IV 4000 recalled. Still haven't put out a working one. Merom delayed till 4th quarter atleast, Intel finally putting out a 64 bit laptop cpu. Lots of paper releases with actual product lagging months behind. Pentium IV extremely expensive and no great improvement editions. Xeon eaten for lunch by Opteron.
The Conroe benchmarks I have seen were very fishy at best. The 'testing' was so controlled that I don't feel the results have any credibility. At best a smoke screen hoping to delay purchases. If the test had any credibility the testers would have been able to look inside the machines, check bios settings etc. Anyone can rig a test to produce just about any results. Let them spend a few days on the machines, look inside, see what drivers are being used etc.
Call me when they do that and then I'll beleave the shit I see. I'm pritty sure that has ya really going there. You would be a fool to beleave benchmarks in such a controlled way of "testing" things. It makes me sick how much of a lie conroe is. And ppl can be suckered enough to beleave how good it is with such questionable credibility.
Uh... just for the record, Intel has had more failures than AMD has had success and Intel still has more money to burn than anyone but that Gates boy...Serge84 said:
I'll rip on all of them given the chance, but if you never take risks, you gain little.
$BangforThe$
[H]ard|Gawd
- Joined
- Nov 5, 2005
- Messages
- 1,672
Serge84 said:
I will give you one thing . Your glass is diffantly half full
DemonDiablo
[H]ard|Gawd
- Joined
- Jun 7, 2004
- Messages
- 1,767
$BangforThe$ said:
well i dont blame him. looking at all of these "previews" are great and all and they give you a good look at what might be to come but they still arent mass production retail products. . i want to see retail, conroe in mind, built motherboards and "reviews" of these products from every pc reviews site on the web before im truelly sold. also while i doubt am2 will bring great strides in improvements it will still hopefully bring something.
with that all said im without a doubt eagearly awaiting for conroe and hope it lives up to all of its hype for my next pc rebuild in august.
$BangforThe$
[H]ard|Gawd
- Joined
- Nov 5, 2005
- Messages
- 1,672
There is little reason to doubt Conroe given the performance of Yonah on the Aopen 975 cipset M/B .
I am more interested in weather or not AMD can drop the price's of the FX60 to around $300 dollars and still maintain a profit . If they can a FX60 @ $300 would be a sweet deal. Slower than Conroe but still a very nice Cpu for $300.
I am more interested in weather or not AMD can drop the price's of the FX60 to around $300 dollars and still maintain a profit . If they can a FX60 @ $300 would be a sweet deal. Slower than Conroe but still a very nice Cpu for $300.