Discussion in 'Hot Deals' started by suraswami, Oct 27, 2008.
It does require some good internet.
Actually, when a cache miss happens, instructions are still being executed. They are just being used to do a different task. So a task that would normally take 1000 instructions(making a number up just as an example), might take 1250 due to the extra instructions needed because of the cache miss.
The IPC of the two processor are the same. This doesn't mean they do the same amount of work, or they are both equally efficient.
IPC and I/s isn't in any way accurate judge of processor performance.
Not as cheap as when I sold my 3800+ am2 to my friend for $10
Meh, good deal but the power usage is to high. That semprom that has heatsink with it and uses 45w for $25 would be decent for a low power build.
Instructions that are being thrown out. IPC counts only include successful execution of an instruction--something that produces a good end result.
Using your example, 1000 counts to successfully executed instructions for both. The time wasted to redo 250 worth of instructions is dead time. Instead of averaging, say 2 IPC, it would only be averaging 1.6 IPC. The architecture sets the theoretical max usually in favor of the ALUs. It is the actual work that gets completed that counts.
It is the only way to judge processor performance. Everything (benchmarks, games, etc.) stems from those figures.
It is the actual work done that counts in the end, but that is not the IPC count. You really need to look up IPC and research what it means.
I get what you are trying to say, and I agree. However, you are making up definitions for terms on your own.
And in my example, the instructions aren't being redone, the extra instructions caused by the cache miss are used retrieving the data that wasn't in the cache from memory. Those aren't wasted instructions, they are still used, they are just used to do the added work required when there is a cache miss.
All processors in the same architecture have the same IPC, that means all processors based on the Core 2 architecture have the same IPC and all processors based on the K8 architecture also have the same IPC. This means the sempron will have the same IPC as an Athlon. However, the processor is not as efficient due to the extra work needed when there is a cache miss, and the fact that the lower processors have more cache misses.
If IPC is the only way to judge performance of a processor, then a Celeron Dual Core would perform identically to a Core 2 Duo as they both have the same IPC. There is far more involved in processor performance than just IPC.
I never looked it up until now and what I see fully supports what I am saying:
Heh, kinda looks like I authored it, but I didn't. XD
Fetching extra data isn't necessarily an instruction--it is the continuation of a previous instruction carried out over more clocks. Again, a deteriment to the IPC figure.
you can get some sick overclocks on them 3+ghz is possible YMMV
Heh, yeah. Small caches are a mixed blessing. They take up less silicone space so they are cheaper to manufacturer and because they are simple in design, they tend to overclock better; however, they also aren't as efficient at completing work, clock for clock.
Indeed, if you believe wikipedia, a site made up of information provided by people that usually have no clue what they are talking about. I would be even more inclined to believe the Wikipedia article if it didn't have This article does not cite any references or sources.
in huge letter at the top of it.
Though even if we go by what wikipedia says, and I'm not saying we should, nothing really there supports anything you have said. IPC can vary even on the same processor depending on what software is run, I don't see how that helps your argument any. Different software uses different instruction sets, this can have a great impact on the IPC for different software on a given processor. It talks nothing of IPCs being different because of the useful work done on the processor or cache misses.
And even if you read futher down you would have seen:
Here is another area you are wrong. In terms of the Celeron and Sempron, they don't take up less silicon space, and aren't actually any cheaper to manufacture than their full cache'd brothers. Celerons and Semprons are the same as their bigger brothers, they simply have sections of L2 cache that are defective. The defective sections are disabled, and the processors are sold with lower cache. They are not any cheaper to produce, they actually cost exactly the same, however they can be sold for much cheaper priced, because if they were not sold, they would be a loss as they would just be thrown out. It is much better to sell a processor for next to no profit(which is usually the case with the Celeron and Sempron) than it is to take the loss of just throwing it away.
Sometimes they do overclock betteer with the smaller cache, as having less cache means less to go unstable. However, the instability has moved away from the cache, and we are starting to see both the high end and low end maxing out at very similar speeds. Though this even varies from processor to processor greatly, even varing greatly in the same stepping(the G0 Q6600's being a good example of this).
One advantage though, is the relatively small power consumption and heat output the lower cache chips produce. Disabling large sections of L2 really helps lower heat production, especially since the L2 takes up the majority of the die.
@Tekie & Ford
All i know is "Celeron" & "Sempron" can be harsh words to use nowadays. Gives the assumption that they aint crap. Have to debunkify those words. Really sit back and look at the specs.
They may not be the greatest but they aren't horrible. Compared to other higher end products they really don't "shine" as well.
Yah know what I mean ..... ? ?
LOL first gaming rig I ever built was that exact $20 sempron (I paid $70 for it back then) that I OC'd to 2.9Ghz coupled with a vmodded and highly OC'd 7600GS 256M and a $50 Gigabyte AGP mobo. Great HL2 rig . Superpi in 31 sec!! I miss that thing. This is a great chip just to mess around with and not worry about frying it.
EDIT: nvm it was that chip but in a 754 flavor.
the sister (or brother) of this Sempron (the 3200) i have had and overclocked the shit out of it. I got 3.2ghz from it and gamed on it with a overcloked 7900GS played everything just as good as th x2 4600+ i had at the time scored around 5k in 3dmark 06 and about 15k in 3dmark 03 and around 28s in Super PI. The 3400+ has higher L2 then the 3200+ so if you could get it to the same speed it should be slightly faster.
Definitely, they certainly aren't the crap that most assume when they hear those words. However, they aren't high end processors by any means. They certainly are good enough for your normal computer user though, anyone that doesn't play games or do much media encoding really.
hey first build was a ECS mobo and Duron 1.6ghz cpu. That thing was freaking fast for my friend who had a P4 2.4ghz. Even Oced to 2.2ghz with better mobo and did all my media encoding too for atleast a yr. It still serves well on my Father's office PC for report generation and data analysis.
Yeah, my first computer I built myself was a Socket 478 Celeron 2.8GHz Northwood, I got that think up to 3.4GHz. I replaced it with a Northwood Pentium 4 3.0GHz.
My first 775 build was a Celeron 352, which I overclocked the snot out of, got it to 4.68GHz. Replaced it with a Pentium D 805, which I also clocked the snot out of. I still have Celeron 352's in office machines at work, they do the job wonderfully(their still folding for TPU too ).
Definately not good enough for 1080p hd video. It would be choppy with 720p HD upscaled to 1080p resolution even.
i can get 1080p smooth off an old P4 1.5GHz laptop (dothan core), and my old media PC was a 2.4GHz P4, that managed 1080p. I'm pretty sure this CPU combined with the acceleration of a video card, can pull it off.
"...memory hiearchy..." (hint hint).
Yes, I know it isn't 100% foolproof just because the software has to be authored in a way that only stresses one instruction; however, we're talking averages here. It is a good way to compare similar processors using similar benchmarks.
Some are butchered versions from the Athlon/Core 2 line--the processors that failed QA testing. Not all are failed processors though (e.g. Conroe-L).
i thought on AMD chips this was somewhat made up for by the on board memory controller? a technology intel hadn't adopted until the release of its upcoming chips. so their must be something to its claimed advantages
Actually according to your article, it isn't a good way to judge performance in any way.
I thought the Conroe-L was just a regular Conroe(or Allendale) with a core disabled. I didn't think the Conroe-L was actually its own core, but I could be wrong.
Separate names with a comma.