CPU FAQ Beginners, Unofficial

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
THx u guys! I PM'd Fredi...but he didn't reply!

I've actually found a few other forums that seem to me more tech orientated than this one...so I don't come around here much anymore. It's a bit slow (man, can it take a while sometimes to load a page) and completely unmanaged. And it's mighty rude for a mod to just delete a post offering a helping hand as if it were spam...


<font color=red>01010011010</font color=red>
 
They don't even do calculations. Only 1 stage out of 20 in the integer pipeline does calculations, all the other stages are either issuing or fetching or decoding, etc.

Decoding & Encoding ARE calculations. Fetching & Issuing require calculations...perhaps the numeral "1" and complete are not to your liking...I can see that argument. It's a figurative term in the sense in which I was using it, not a literal breakdown. So just spit out your better answer...much more briefly and clearly than previously, please.

And I fail to understand how your argument with the naming technique is valid. Perhaps you should rethink that argument. You say that one is commercial, and the other is code? That makes no sense to me, as they would both be commercial if they're released to the public. Not to mention where did I mention code or commercial? And the prefixes that Intel does use are merely their abbreviations they use of the same notation (though not really much of a resemblance of the larger name). So you're saying that even though they (the single letters and longer names) denote the same part(s) of the CPU, they're not the same? And even in your last sentence you note that AMD denotes the Clawhammer structure...It's not just plain "Athlon", is it!?

I would have to say that just from your handle alone, your head is a bit to thick for intellectual argument. Perhaps a leather couch environment would be better for you to debate in. To be honest, I'm only arguing with you because I enjoy debate...your arguments are very weak, but do have a few points that some other readers may inadvertently question (even though you're not very clear either).

Perhaps you would prefer the word "Process" be used in place of "Calculation"? Would you argue against this? I think you could have shortened your argument by just stating a change in the "1" and "calculation" need be made for a slightly more accurate description. As for complete, one pass of the all CPU's functions is as complete as it gets. Complex calculations may require more usage.

Why did I even respond to this argument? I must be bored...and feel like rambling, i guess.

<font color=red>01010011010</font color=red>
 
Decoding & Encoding ARE calculations. Fetching & Issuing require calculations...perhaps the numeral "1" and complete are not to your liking...I can see that argument. It's a figurative term in the sense in which I was using it, not a literal breakdown. So just spit out your better answer...much more briefly and clearly than previously, please.
The process of fetching and decoding go through a logic die that doesn't do any real math, which is what calculations are.

And I fail to understand how your argument with the naming technique is valid. Perhaps you should rethink that argument. You say that one is commercial, and the other is code? That makes no sense to me, as they would both be commercial if they're released to the public. Not to mention where did I mention code or commercial? And the prefixes that Intel does use are merely their abbreviations they use of the same notation (though not really much of a resemblance of the larger name). So you're saying that even though they (the single letters and longer names) denote the same part(s) of the CPU, they're not the same? And even in your last sentence you note that AMD denotes the Clawhammer structure...It's not just plain "Athlon", is it!?
The difference between commercial and internal code name is pretty obvious. One is the official name of the thing and the other is what Intel engineers and possibly even market people call it before it is released. "Athlon" is a patented, commercial name. AthlonXP denotes that it is the single-processor palomino line. Intel chose to denote their processors with suffixes "a" and "b". In absolutely no marketing campaign has the word "Northwood" or "Palomino" ever been mentioned. Nor does either company use them to market to the consumer. The "Athlon" name has been used for both the "Athlon classic" and "Athlon t-bird". AMD chose to give it a new marketing name for the Palomino line because it felt it needed a boost in marketing. There is no set reason for "hey, the core's changed, maybe we should change the name". Pentium 4 is still the official name for the P4 processors. Celeron has been the official name of the budget processors from Intel since the days when it was based on the old P2 design. The marketing name has not changed since.

I would have to say that just from your handle alone, your head is a bit to thick for intellectual argument. Perhaps a leather couch environment would be better for you to debate in. To be honest, I'm only arguing with you because I enjoy debate...your arguments are very weak, but do have a few points that some other readers may inadvertently question (even though you're not very clear either).
And yet you're the one to pick up insulting me. I enjoy nitpicking once in a while as I indicated what I was doing in my first post. Why you even care is beyond me.

Perhaps you would prefer the word "Process" be used in place of "Calculation"? Would you argue against this? I think you could have shortened your argument by just stating a change in the "1" and "calculation" need be made for a slightly more accurate description. As for complete, one pass of the all CPU's functions is as complete as it gets. Complex calculations may require more usage.

If you'd read my whole post you would've seen what I said about modern hyperpipelined MPU's not actually doing one whole task in one stage of the pipeline. There are 2 pipeline stages in the P4 that do nothing more than just to relay the signal. So no, each "clock" does not do 1 thing. The only stage that is always guaranteed to do 1 complete thing is the execution stage.
<P ID="edit"><FONT SIZE=-1><EM>Edited by imgod2u on 07/12/02 08:58 AM.</EM></FONT></P>
 
heh heh

I thought it was fun...did you? And, since I was "lacking", you must now explain all the pipelines for every cpu that uses a MHz rating, and exactly what pipelines have been used in the rating, as you only have mentioned the P4. And you say that decoding & encoding are not calculations? A calculation, in my opinion as fact, is not merely the completion of a mathematical equation. It is:

Main Entry: cal·cu·la·tion
Pronunciation: "kal-ky&-'lA-sh&n
Function: noun
Date: 14th century
1 a : the process or an act of calculating b : the result of an act of calculating
2 a : studied care in analyzing or planning b : cold heartless planning to promote self-interest

and as for calculate:

Main Entry: cal·cu·late
Pronunciation: 'kal-ky&-"lAt
Function: verb
Inflected Form(s): -lat·ed; -lat·ing
Etymology: Latin calculatus, past participle of calculare, from calculus pebble (used in reckoning), perhaps irregular diminutive of calc-, calx lime -- more at CHALK
Date: 1570
transitive senses
1 a : to determine by mathematical processes b : to reckon by exercise of practical judgment : ESTIMATE c : to solve or probe the meaning of : FIGURE OUT <trying to calculate his expression -- Hugh MacLennan>
2 : to design or adapt for a purpose
3 a : to judge to be true or probable b : INTEND <I calculate to do it or perish in the attempt -- Mark Twain>
intransitive senses

I see more than just mathematical computation in there for a definition that applies to many of the functions of a cpu, eh? I consider logical functions to be types of calculations, but I see you do not. We also have analyze, planning (fetching & prefetching), solve or probe the meaning of (encoding & decoding), to judge to be true (logic), to design...I just happen to agree with Webster.

As for the marketing, you seem to think that only one marketplace exists for the computer hardware world. Are we not of a different market than that of the uninformed BestBuy'ers. No, they don't plaster all the names on their packaging and posters. They let us savvies use our own marketplace...makes us feel smarter...which we generally are. We use their "code names" in the guru marketplace (after all, do you not see them when ordering from so many vendors? and if not, I bet you might ask!).

Man, I was just going to reply with a "😛PPPP", but darn if I didn't ramble again...but one last thing. So what exactly are you saying? Really...what???

There is no set reason for "hey, the core's changed, maybe we should change the name".
kind of just happens...no, they don't usually change the "general consumer" advertising, because the general consumer is already pretty lost in the computer world. But if they're not changing the name...then...???? Thunderbird, Palomino, Coppermine, etc,etc,etc??? Yes, there's just a pattern, eh...no reason?

smoke a cig, you'll feel better


<font color=red>01010011010</font color=red>

[EDIT]

Man, what planet are you from??? Where do you see anything in my post(s) about marketing? I'm not going to argue with an idiot as such any further...you're wasting time and forum space. I'd rather see pointless posts than a pointless post that thinks it's an omniscient post. Oops...that sounded like a FatBurger comment...I retract back to "...wasting time." Scratch the latter. Ya know what...scratch the idiot thing to...I actually think I'm enjoying this argument. But ya do go a bit of course in your argument...
 
I'm gonna have to play the evil guy here, cuz I should add that a local shop here advertised in a magazine the core names in BIG!
One was "AthlonXP with the PALOMINO core!!!" Other was "New Pentium 4 with Northwood Core!!!" Or something like that, it was pretty hyped though.

--
:smile: Intel and AMD sitting under a tree, P-R-O-C-E-S-S-I-N-G! :smile:
 
I see more than just mathematical computation in there for a definition that applies to many of the functions of a cpu, eh? I consider logical functions to be types of calculations, but I see you do not. We also have analyze, planning (fetching & prefetching), solve or probe the meaning of (encoding & decoding), to judge to be true (logic), to design...I just happen to agree with Webster.
And if you'd read my post, you would've seen me mention that each stage in hyperpipelined MPU's don't even do 1 complete task. So even if a task such as decoding does count as a "calculation" as you seem to prefer to make this a symantec arguement, 1 stage STILL doesn't do one whole calculation. Not to mention the 2 stages in the P4 that don't really do anything but relay the signal.

As for the naming. You said Intel and AMD included these suffixes (or was it prefixes you called) to indicate small changes. In order to be a suffix or prefix, they'd have to be in the same catagory. IE, Mustang GT are both marketing terms. If GT was simply an internal code the engineers used (you don't see anyone saying "whistler" is a suffix to WinXP now do you?), it would not be part of the name and would not be a "suffix". Whether individual vendors choose to use the internal code to market or not is irrelevent. They are not "suffixes" or "prefixes". The code-names have absolutely no relation to the marketed names.

<P ID="edit"><FONT SIZE=-1><EM>Edited by imgod2u on 07/12/02 04:04 PM.</EM></FONT></P>
 
And if you'd read my post, you would've seen me mention that each stage in hyperpipelined MPU's don't even do 1 complete task. So even if a task such as decoding does count as a "calculation" as you seem to prefer to make this a symantec arguement, 1 stage STILL doesn't do one whole calculation. Not to mention the 2 stages in the P4 that don't really do anything but relay the signal
Yes, but the MHz only applies to how complete of a calculation the chip is capable of making is one stroke, or pass. It may require more than one Hz, or cycle, for the CPU to complete the necessary calculation, but that is part of the POWER of the chip, not the speed. Since you bring cars into it, MHz is the Horsepower, and the Power of the CPU is the torque. You won't be able to complete one trip in your car on one cycle of the cam, unless maybe the torque is so ungodly it instantly rockets you to mach1. But that doesn't make the completion of all cyclinders firing any less complete, eh?

And the windows thing...we are talking about CPU's...but still, let's give an example. The Pentium 3: Coppermines are denoted with an "EB" suffix, Katami 133Mhz FSB are denoted with a "B" suffix. As for the GT, is too is an abbreviation for a very large list of differences in the product. I don't understand what you mean by they're not in the same catergory. Again you throw the "marketing" in there. We're talking technical, not general advertising & marketing...quite the difference there is. My intial post was relating the naming techniques to child & family naming here in America...not marketing. And just because AMD doesn't launch a massive television ad campaign with the core names mentioned, does not mean they do not exist or have nothing to do with the CPU.

<font color=red>01010011010</font color=red>
 
It would be great to address the ever popular AMD vs Intel question.
And what is that question? Which is better? I really don't think there's much difference. Our PC power, though quite empressive to us consumers, is still relatively weak. I don't think there is a whole lot of difference between the two. AMD does have a more powerful chip, but Intel has taken the lead in speed. Most of the differences are personal opinion, having to due with their loyalty of having already dealt successfully with one, and perhaps not with the other. But they're both equally compatible these days. I prefer AMD because I have dealt with more AMD systems. But If AMD just all the sudden dissappeared, I wouldn't mind a P4...in fact, I wouldn't mind one anyway just to have a good variety.

I'm pleased with both companies and chips...and you HAVE to have competition in this world to ensure progress and quality. But that's just my $.02

And I'm glad to see dealers/vendors using the core names for marketing. Our user market needs some serious training in the art of computers (anyone in tech support will back that one up). Now if we could just get some requirements to complete some sort of course/training/tests before they're allowed to even speak with a technician...

j/k...that would never work.

<font color=red>01010011010</font color=red>

[EDiT]

on an added note...I wonder how long this debate will ensue. (crap, no spell checker on this PC...doh!)
 
Yes, but the MHz only applies to how complete of a calculation the chip is capable of making is one stroke, or pass. It may require more than one Hz, or cycle, for the CPU to complete the necessary calculation, but that is part of the POWER of the chip, not the speed. Since you bring cars into it, MHz is the Horsepower, and the Power of the CPU is the torque. You won't be able to complete one trip in your car on one cycle of the cam, unless maybe the torque is so ungodly it instantly rockets you to mach1. But that doesn't make the completion of all cyclinders firing any less complete, eh?
I believe your exact statement was each "Hz" completes one calculation. Which is not true. As for the ability of each stage to do work, that's really a part of the average IPC.

And the windows thing...we are talking about CPU's...but still, let's give an example. The Pentium 3: Coppermines are denoted with an "EB" suffix, Katami 133Mhz FSB are denoted with a "B" suffix.
And the Coppermines with 100MHz FSB was denoted with the "E". Both "E" and "EB" are coppermines. The "coppermine" was suppose to correlate to the "e" suffix, but it is not a suffix in of itself. Intel used the "e" suffix as an indication in the model number, never mentioning Coppermine. That is merely the name the engineers use.

As for the GT, is too is an abbreviation for a very large list of differences in the product. I don't understand what you mean by they're not in the same catergory. Again you throw the "marketing" in there. We're talking technical, not general advertising & marketing...quite the difference there is.
I'm making the true distinction between the "Coppermine" and "Northwood" names and the commercial "Athlon" and "Pentium" names. That's what they really are. This suffix thing you brought up is total nonsense. In no way was "Coppermine" ever a suffix or prefix for the "Pentium 3" line used by Intel. It was merely something the enthusiastes liked to throw around to sound smart (and avoid explaining the change with every mention).

My intial post was relating the naming techniques to child & family naming here in America...not marketing. And just because AMD doesn't launch a massive television ad campaign with the core names mentioned, does not mean they do not exist or have nothing to do with the CPU.
I said they have nothing to do with the commercial name, not the CPU. And as I pointed out, that is the main difference between "Athlon" and "Palomino". Athlon is marketing, Palomino is what the engineers use. THAT is the difference. Whether you want to talk about marketing or not is your choice. But the correct answer to what is the difference between the "coppermine" name and "pentium 3" name is that one is the internal code and the other is the marketed name. THAT is the difference. They are not suffixes or prefixes. The "a" and "b" and "e" are suffixes.
 
In reply to:
------------------------------------------------------------------
In reply to:
--------------------------------------------------------------------------------

It would be great to address the ever popular AMD vs Intel question.



--------------------------------------------------------------------------------

And what is that question? Which is better? I really don't think there's much difference. Our PC power, though quite empressive to us consumers, is still relatively weak. I don't think there is a whole lot of difference between the two. AMD does have a more powerful chip, but Intel has taken the lead in speed. Most of the differences are personal opinion, having to due with their loyalty of having already dealt successfully with one, and perhaps not with the other. But they're both equally compatible these days. I prefer AMD because I have dealt with more AMD systems. But If AMD just all the sudden dissappeared, I wouldn't mind a P4...in fact, I wouldn't mind one anyway just to have a good variety.

I'm pleased with both companies and chips...and you HAVE to have competition in this world to ensure progress and quality. But that's just my $.02
-----------------------------------------------------------------
The "Which cpu should I get?" is a FAQ.



"Just the facts ma'am"
 
The "Which cpu should I get?" is a FAQ.

I don't know if I would be able to answer that question in a manner that would apply to all, or even the majority, of the consumers round yonder parts.

As for Impathetic2u...you're WRONG. Get over it...bring up some right points, and the argument will continue. Too many incorrect points to continue to post...


<font color=red>01010011010</font color=red>
 
"As for Impathetic2u...you're WRONG. Get over it...bring up some right points, and the argument will continue. Too many incorrect points to continue to post..."

You're a real sore dude. I think ImGod had a good argument. You dont agree with him so you get viscious.....not cool man. For the record, fetch and prefetch arent calculations, they're setting up info to be digested. Each stage isnt a calculation, it's part of a calculation as a whole. The Athlon "family" IS just marketing, and not a "family" of similar cores as you stated.....The Hammer isnt a little core change.
Calling someone names is pretty easy eh?

This sig runs too hot.
 
This post is for everyone.I dont seem to find a drop down menu with all the available posting options that some other boards have.At any rate,this discussion reminds me of something at the Anandtech site.Way too many attitudes.
Someone took the time to make a F.A.Q that they thought might be helpful to some,and immediately the wolves attack.While the document might not be 100% technically correct,it does supply enough information to help out a "newbie",and if memory serves,he initially stated that this WAS the target audience. Furthermore,there was no intent in the original post to "dazzle" us with brilliance,nor to "talk down" to the masses.For all intents and purposes,it was just an attempt to provide a useful tool that was met with a "piss poor greeting".He has every right to be upset.If you dont like the quality of the document,then write better one, or be quiet.People that like to criticize,but are too lazy to do something themselves,make a forum look real bad,real quick.....🙁

If ya don't ask..How ya gonna know.
 
yeah i agree with vetplus40. i started reading a nice post for beginners and ended up getting my eyes crossed trying to figure out what you guys were talking about! sure, there were a few errors, i guess, but it was a good FAQ and i wish i would have seen it earlier. i suppose it got buried and i overlooked it a few times...

pekstein

<i> Garth, marriage is punishment for shoplifting in some countries. </i>
 
Nobody ever said it wasn't a good FAQ. I nitpicked some things that weren't correct. It started out with the poster calmly disputing that his post was indeed correct. After a while he turned bitter either because he realized he was wrong but just didn't like it being pointed out or is truely that arrogant. I don't know. I would've thought he would welcome some additional info to improve his FAQ.

<P ID="edit"><FONT SIZE=-1><EM>Edited by imgod2u on 07/14/02 10:42 PM.</EM></FONT></P>
 
Sometimes the bigger issue is not what you say,but how you say it.Positive correction is generally better received than "nitpicking".I understand that you were just trying to be helpful.I also understand that incorrect information need to be corrected.Sometimes what we say is interpreted in a different manner in which it was intended.This appears to be the case with this post..:)

If ya don't ask..How ya gonna know.