You can also assume the exact opposite - Intel may be slacking a little since they are way ahead of AMD. By showing off great benchmarks Intel might not feel so safe anymore and make a greater effort. By intentionally leaking crappy benchmarks you might fool Intel into slacking because they feel über-safe.
Slacking for all of 3 months? Sorry, it don't work like that.
The primary purpose of "leaking" performance expectations which suggest next business quarter's product line will trump your competitors product line is to attempt to influence those discretionary purchases and upgrades whose timeline are not set in stone by the budgetary powers that be.
Every XEON purchase that could be delayed for sake of uncertainty in the K10 performance/power expectations is a potential new sale for AMD that would have otherwise went to Intel.
The ONLY reason you don't leak performance expectations is because the only one's you got are unflattering of your product and will do little to disuade XEON purchases set to occur in the current business quarter.
This is how marketing works. Disagree with me all you like, I really don't care how ignorant you remain, but it is reality nonetheless.
Well that makes sense and it it's probably related to what's been going on.
On the other hand, alll the "rumors/leaks" of early samples on beta boards has made Phenom seems worse than its really going to be - and still AMD hasn't released muchr eally to counter that (that 3GHz thing, and that was about it). If your product is better thant what its rumored to be, wouldn't they do something about it?
Point being, it's a lot more complex than the simple scenarios we painted out.
On the other hand, alll the "rumors/leaks" of early samples on beta boards has made Phenom seems worse than its really going to be - and still AMD hasn't released muchr eally to counter that (that 3GHz thing, and that was about it). If your product is better thant what its rumored to be, wouldn't they do something about it?
Point being, it's a lot more complex than the simple scenarios we painted out.
IMO the lack of information on K10 reflects the internal chaos at the company. They just bought ATI, and they're working on a bunch of projects all at once. I don't think K10 has had top priority since the merger; ATI was purchased to build Fusion.
Since K10 was lower priority, I guess it didn't have a solid release date as it was relying on whatever resources were leftover from Fusion. AMD was probably wary giving out information on it as it would have been far too preliminary. They did give out some performance figures (40% faster floating point or something), and showed a 3ghz setup, so I guess in a way they're trying to hype the chip a bit.
Really it seems like AMD has too many projects on the go to really focus properly on anything. Even after Barcelona there's Bulldozer, Fusion, and whatever ATI is working on (chipsets and graphics). If you look at intel, they're 10 times bigger than AMD and are pretty much focussed entirely on creating the fastest desktop and server microprocessors. Sure they do chipsets, but the resources required don't have the effect that they do on a smaller company like AMD.
Since K10 was lower priority, I guess it didn't have a solid release date as it was relying on whatever resources were leftover from Fusion. AMD was probably wary giving out information on it as it would have been far too preliminary. They did give out some performance figures (40% faster floating point or something), and showed a 3ghz setup, so I guess in a way they're trying to hype the chip a bit.
Really it seems like AMD has too many projects on the go to really focus properly on anything. Even after Barcelona there's Bulldozer, Fusion, and whatever ATI is working on (chipsets and graphics). If you look at intel, they're 10 times bigger than AMD and are pretty much focussed entirely on creating the fastest desktop and server microprocessors. Sure they do chipsets, but the resources required don't have the effect that they do on a smaller company like AMD.
plasmabomb is the only one correct here.compression goes down as rpm increases.cranking compression on an engine is about 150-180 psi.running compression same engine is 80-90 psi @700 rpm.1500 rpm-60 psi.superchargers provide boost at around 15-18psi.if you tried just 50 psi youd blow the engine,not even coming close to compression readings at crank.the reason it blows is too much air/fuel for a given volume.
same principle as a rifle cartridge,too much powder it blows up in your face not out the barrel.nothing to do with compression.diesels have much higher compression but they dont blow till you put a low octane fuel like gasoline in that burns much faster.you have your terms and physics wrong.Btw im a performance auto tech for more than 35 years,all facets
rotary to diesel,as well as flex fuel, natural gas,propane,hydrogen.Believe me i know what im talking about.
just noticed my sig man its been a while,the only rig still running is my kids nf7s,gotta update that.
"Boost increases with the square of RPM (unlike the linear nature of the positive displacement devices), however low-rpm boost suffers due to the fact that air can pass back through the supercharger with little restriction until RPMs rise sufficiently to counteract the effect."
Link
"This makes the maintenance of smoothly increasing RPM far harder with turbochargers than with belt-driven superchargers which apply boost in direct proportion to the engine RPM."
Link
"Normally the supercharger only begins to create boost at around 3000 rpm, and the boost curve gradually and increasingly rises with engine RPM."
Link
"This is especially true of centrifugal supercharging, which generates boost in line with engine rpm"
Link
The general consensus appears to be that superchargers generate compression in proportion to RPM. But we are veering off the original topic here - the reason CPUs cannot be compared with engines, much less supercharged engines.
Nemesis 1
Lifer
- Dec 30, 2006
- 11,366
- 2
- 0
Pure fud. First were talking about performance engines. Most 11.5 to 1 compression is about 225 cranking . The engines I am familiar with are 13-14 to 1 compression ratio which comes out to around 275-300 psi. cranking . Now running compression is very dependent on the performance camshaft. Back in the late 60's early 70's . Camshafts ran a lot of overlapp. which does effect running compression but nothing like what your talking about. Todays high performance cams in a small block chevy on carb make anywere from 650-to 850 hp depending on component selection . In 1970 getting a small block chevy to make 500 hp . was very hard work. Heads and cam tech changed all that.
So tell me this . If cranking compression is 300psi . and the valves and cam stay at the same position all the time were is all this compression disappearing to when engine is running. everthing is the same @ 100 rpm as it is @ 8500 rpm . If the carb. is wide open when spinning the engine at cranking (100rpm) . When running the carb. is at full throttle 8500rpm were does this magical loss in compression go. Everthing is = except rpm . Plus the fact that a running motor develops what is called vacum . Which helps to pack the cylinders. Now if you have performance engine running open throttle you may see 1 bar of vacum . In a carb . their is whats called a power valve . this power valve works off the vacum of the engine. lets say you have a 5.5 power valve. that valve will not open until you go below 5.5 bars on vacum gage which feeds more fuel to the intake system . Now the moment you close the throttle vacum will instantly shot up to about 8 bars. Than that power valve closes giving less fuel to the system .
I would very much like to see alink to your 180 psi cranking going down to 90 psi while running. Link please.
Than your statement about overpacking a cylinder is really really funny. The older guys here would understand. . The reason that the engine blows has nothing to do with to much fuel air but has everthing to do with detonation.
The old guys here can remember the 60's muscle cars. If you put cheap fuel in it the motors would knock and when you shut them off it would run after shutting off the ignition dieseling. Ping is what many called it . My poor misguided friend . This is a result of to low of octane @ a given compression ratio . ignition timing has much to do with it. Detonation is what blows motors not to much fuel air. Nascar has taken compression up to 16 to 1 and thats about as far as you can go on the best racing gas money can by. Than there is methonal which can go to about 18 to 1 . Than you get to nitro in the big drag cars with blowers packing those cylinders full . To much blower power(To much belt drive gear) will blow the engine more because of component failure do to excessive HP. Detonation is poor fuel or to much compression for the fuel type used or ignition timing is advanced to far . Or air fuel ratio is off by a bit.(lean burn cause detonation not enough fuel ) To much fuel causes excessive cylinder ware or plug fouling
So now that you have your links turbochargers use waste gates to get rid of excessive cylinder chargeing. I had a 88 turbo t-bird brand new. 4 cylinder 143mph. topped out at. Boost pressure was as i recall was 6 psi. Than the waste gates opened so as not to cause detonation . From you got it right poor fuel . Has nothing to do with how much fuel . because air fuel ratio is pretty much a constant.
Not really sure how belt driven blowers got in here but there a whole nother type of system . that pack cylinders at given rpm this ratio is determined by the sizing of the belt gear drive.
In 90 i traded my turbo coupe for a super coup belt drive blower again about 6 psi and it was a V6. The turbo coup was better in all areas driveability top end performance and acceleration. Love those waste gates.
Not that many years ago on a 600hp sb . 38degree ignition timing was considered about right. Today its 34 degrees or lower funny how things work. Is whats really funny is why they have gone down in ignition timing while raising HP. But I won't discuss that . For good reasons.
Please, everyone remain on topic. Internal combustion engines can be discussed in "The Garage".
Thanks - Anandtech Moderator
Nemesis 1
Lifer
- Dec 30, 2006
- 11,366
- 2
- 0
Yes it was but a member used a performance engine as a comparsion to cpu and was shot down . I on the other hand agreed. but for differant reasoning . I used waste gates on a turbo boost as a comparision to core issue in this case C2d 4-5 issue vs K10 3issue. The waste gate pressure or core issue for Intel C2D is much higher than K10 . Which in c2ds case = more processing =higher performance vs. K10 . In this case K10 has a much better chance of bottle nicking do to its 3 issue core than Intel has with it 4-5 issue core. C2D burns its fuel lean . While K10 burns its fuel rich which could cause a fouled plug in k10 . I am sure That fuel air ratio would have been better terms to use than waste gates. Waste gates would have been better to compare to like l1 or l2 cache or other logic units in the cpu .
To say cpus and engines are not apple to apple is a no brainer or is it . If one considers one thing common to many things. FLOW! C2D will out flow K10.
Bandwidth is = to cubic inches If you don't have FLOW all that bandwidth is wasted.
To say cpus and engines are not apple to apple is a no brainer or is it . If one considers one thing common to many things. FLOW! C2D will out flow K10.
Bandwidth is = to cubic inches If you don't have FLOW all that bandwidth is wasted.
Mergers and acquisitions (M&A) don't really happen like that. Re-orgs do.
In a M&A the pre-existing business groups tend to motor along unaffected by addition of the acquired business groups. The K10 group existed before ATi was brought in-house and will to a large extent carry out there charter mostly uninhibited.
The next-gen stuff, where Fusion changed the roadmap, those are the pre-existing AMD groups that no-doubt got re-organized after the M&A activity.
I admit from the consumer side of things you get to see so little of what goes into making a successful product (from concept, to design, to mfg, and distribution/sales/support) that it can appear that something seemingly huge like the M&A of ATi ought to be impacting the K10 timeline but I assure you K10 was way to far down the pipeline to be impacted by the M&A with ATi.
Throughout the entire prototype and pre-production (as stated in my last message) process, certain features on the CPU, in the BIOS, or on the chipsets have been turned off/on, latencies have changed, etc, etc. This is a normal part of the engineering process as the design is fleshed out and finalized. It does not represent final silicon capabilities and performance.
As I said earlier, I used a poor example as it was not meant to be taken literally spec for spec when comparing engines and CPUs. Regardless of the example, the point was that the platform performance improved significantly as the core speeds improved and this included performance per watt among other indicators. There is a myriad of reasons as to why this occured but considering the early silicon, BIOS, and chipset designs, we could only speculate as to why and I tried to present a few reasons that we honed in on.
If you compare a B00 chip from May to a B02 today, there is a significant difference in performance in all areas (26 seconds in SuperPI 1m for one) and my comments represent observations of what has occurred over this time period. We have final silicon now and results will be posted in the near future. My observations today are different than they were two weeks ago and as the platform matures they will change again.
Once we see the HT 3.0 capable chipsets and Phenom cores mature then we will have an even better indication of the performance of this core design in the consumer market but for now the initial release is Barcelona in the enterprise market.
Nemesis 1
Lifer
- Dec 30, 2006
- 11,366
- 2
- 0
I agree with you 100% Gary. So would it be safe to assume that A0 steppings of Penryn on a bran new process type and logic as well as softerware improvements in Penryn stand to gain as well? Beings how its AO stepping I would say large improvements. Or do you think intel showed their whole hand in this poker game.
I have heard many say AMD is sandbagging . I find that very hard to believe considering there financial position . If anyone is sandbagging here I am thinking its Intel. the 9 & 10 of this month is going to be eyeopening for many I suspect.
I have heard many say AMD is sandbagging . I find that very hard to believe considering there financial position . If anyone is sandbagging here I am thinking its Intel. the 9 & 10 of this month is going to be eyeopening for many I suspect.
No...they are releasing stepping B2 or B3 (I believe B3). B1 was from April...
This is the reason for the 6 month delay, they've created at least 2 more steppings since then.
Firstly, the A0 Penryn steppings were the ones they first taped out (that's true for all A0 steppings)...none of the "leaked" Penryn benches have been run on them though it did actually boot.
Secondly, I don't see what AMD's financials has to do with their strategy...BTW, you DO know that they are receiving an extra $700 Million+ this quarter over and above their sales?)
German Subsidy
Russian Deal
Then $700 million loss the next? Wonderful
Nemesis 1
Lifer
- Dec 30, 2006
- 11,366
- 2
- 0
Viditor give it up . I want to see amd k10 50% faster than Penryn clock for clock. Because it has to be to compete . Its just not going to happen . You can scale only so far. K10 is the end of 3 issue cores.
Merom design will reach its maxamum design with Nethalem C (or what ever they call it.
None of that makes any sense.
1. There is no 'pressure' within a processor.
2. AMD's three-issue/retire architecture is not a bottleneck - there are far more factors that determine performance than issue rate alone. Beyond three issue, there are diminishing returns due to instruction parallelism.
2. Core 2 can decode more instructions per clock, but it still only has three ALUs and three FPUs - the same as K10.
3. K10 has double the intruction fetch bandwidth over Core 2, meaning its decoders are more likely to be fully utilised when executing large instructions.
4. I fail to see how you can compare or attempt to draw any parallels between the fuel/air ratio of an engine and the issue rate of a processor. The mind boggles.
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