Jukka Larja (roskakori2006@gmail.com) on 9/27/09 wrote:
---------------------------
>AM (myname4rwt@jee.male) on 9/26/09 wrote:
>---------------------------
>
>>>1) Out-of-order adds more transistors[1]
>>>2) More transistors adds power
>>>3) More power equals lower clock
>>
>> No, higher power doesn't *necessarily* mean
>> lower clock in general.
>
>I wasn't trying to imply that (depending on what you mean by "in general". Now
>that I think about this, Finnish language has two quite different meaning for words
>"in general". I'm not sure whether the same is true in English (I think it is) and
>whether or not I have myself used the both meanings in this discussion)[1]. But
>if we are in everything else being equal land, higher power is not acceptable.
>
>You should see my reply to Seni and reply to that if you deem it necessary.

Seni was spot-on btw; your (2) can be true in many situations, but not in general (e.g. caches which increase transistor count can reduce power consumption), your (3) can be true in particular cases, e.g. for comparison of similar devices, both thermally or electrically constrained, but not in general, when the opposite is often true as mentioned (e.g. Vdd is adjusted to make full use of available cooling solution or supply).

So, in short: cycle time is often a design target; an OoO machine doesn't have to pay clock-rate penalty compared to [an otherwise similar] in-order machine, even if it has way more transistors and consumes way more power. There's nothing about OoOE that imposes such hard constraint on design (at least nothing I'm aware of).

>[1] In general as in "in every case, at least this is true" or in general as in
>"if this, then that, ignoring everything else". I may have used these inaccurately,
>since in Finnish the first one is hardly ever used outside mathematics.
>
>> So why don't you show the theory that formed
>> your understanding above (I'm asking because
>> you had referred to theory)?
>
>Well, it would be mostly basic physics like you can't do work without loss of energy[2].
>I don't think it would be any good for me to list that sort of theories, especially
>since I don't know their names in English (probably wouldn't remember most names
>in Finnish either). It would be a lot of work and as far as I can see, wouldn't
>address the problem, since you probably aren't disputing that sort of theories.

What's the connection with energy? Cycle time is a function of, among other things, #gates b/w the latches (roughly speaking), and the latter is something that's determined by design.

>[2] I may have misused terms work and energy here. Sorry if I did. I haven't really studied physics in English.
>
>>No, what you say is pretty far from being
>> established truth.
>
>I think that is because you think I'm saying something else than I am trying to say.

Well, I'm not trying to figure out a hidden or different meaning in what you say. Is that really the case?

>> (where are the examples I provided, btw? Deleted?)
>
>I didn't think they were relevant. I don't remember exactly what I deleted, but
>it was probably something I didn't disagree as such or maybe didn't have a relevant opionion about at all.

I recall I gave you some practical examples to show that what you seem to think is true is really not.

>>>[1] Not in general case, of course, but in everything else being equal case. You
>>>can always pick horribly complex in-order core, throw most of it away and add simple
>>>OoO to what is left, if you want make that point.
>>
>>Yes, >xistors and >power. But >power is not necessarily lower speed.
>
>Well, if you aren't allowed to have greater power, and the clockspeed is the only
>variable you can adjust, I don't see what else to do.

Actually, you not only can, but should adjust Vdd along with it. Sure, you can crank the clock & Vdd up or down to adjust any CMOS circuit to operate at certain power level and corr. clock, the point is that with such ifs as above, you're changing the game very significantly. In case you argue it makes comparison more fair, then how about we factor in the time to complete a task and thus measure energy spent? This is as reasonable thing to do as allocating the same power budget. IOW, there are many different things which can be specified additionally and lead potentially to completely different answers compared to the original setting.

>By the way, why did you separate this footnote from my list of arguments? They
>only make sense in very narrow context and responding to them without taking that in account doesn't make any sense.

Well, finally some significant steps forward! :) If you want to specify the context, then this is entirely up to you. Don't blame others for not being able to read your thoughts or uncover the real meaning you put in when it's far from obvious or changes the problem significantly.

>> A fanboy of theory is how you characterized
>> yourself, so what's the problem with that?
>
>The fact that you flaunt it around like it should somehow degrade me. Which, of course, degrades me.

Apologies then, I only used it because you did.

>> If you're not going to reread the thread, not
>> going to refer to sources, what do you seek in
>> this discussion then?
>
>I seek to make you understand that you may occasionally read people wrong and not
>every argument you see is actually made. You have a very aggressive style of arguing
>which makes me feel that if I just walk away, you will remain with false impressions
>and diminished as a whole. I was a teacher for a long time, so it's probably in
>my character not to walk away if I think someone has misunderstood me or something else.

Okay then, please fix all of my misunderstandings. :) Btw, if you consider my style very aggressive, you must have missed far "merrier" times on this board. Mind asking what you teach?

>> Again: do you seek answers to your questions or
>> want to hear something from me? I have already
>> told you my view.
>
>Usually, when I end a sentence with "?", I seek for an answer. However, I think
>we should just focus on the three statements at the beginning of the message. If
>we can't resolve those, there's no point trying to resolve something that at least
>from my side of argument stands right on top of those.

If you *really* seek insightful answers, you should contact Hans as I suggested to you many posts ago, for it seems he no longer attends the board. I don't know anyone here who might know GPU subject better than he does.

>> Interesting. I didn't measure pipeline length
>> myself, and I gather neither did Jouni. However,
>> only one of us was making statements which were
>> repeatedly found to be incorrect in this
>> discussion. Who do you think it was? :)
>
>I think for you it was clearly Jouni. I'm more inclined to go with "misunderstood"
>than incorrect, although neither alone is probably enough to explain everything.

Read the thread, really. One of his fantastic arguments was that the last Intel's CPU on 90nm process was Prescott. To make this statement true, "last" must be replaced with "first". Must be one of my many misunderstandings. :)

>> What you say is very far from being "basic truth".
>
>Think about it this way. I think that the differences we have can be summed in
>the three points at the top of this message. They may not be perfectly formulated
>and may take few iterations before we can actually agree what they mean. If after
>that we still have a disagreement, it should be about something I would call a basic truth.
>
>I have given many examples of things I think you were contradictory about or going
>against basic truths (as in what I mean with the three statements). Your responses
>have been something between "but that's just the opposite of what I said" and "I
>don't understand, aren't you contradicting yourself" (not real quotes, just examples).
>I don't think bringing these up again would help.

I don't recall you pointing out a *single* pair of contradicting statements from me, I think it was me who asked for clarif. once -- smth. about OoOE being a waste.

>> Actually I didn't call *you* a bigot, I simply
>> stated what they tend to do (more detailed
>> explanation can be found in dictionary).
>
>Yeah. You just said that factless discussion is a territory of bigots and that
>I'm having a factless discussion. Sorry, but I don't see the difference.

Well, since you referred to theory, I asked you for some references many times, only to see nothing. You shouldn't refer to "theory" when expressing your opinions or views, that's a bad practice that hardly anyone respects.

>> If you want a fact-based discussion, then
>> contribute something; if you're a theoretical
>> fanboy, show some studies; if you're bigot,
>> this is not necessary.
>
>1) People often misunderstand each other
>2) 1) is even more likely when one or more of the participants are not communicating in their native languages
>3) People are often blind to their own mistakes
>
>Now, I don't have a study at hand to prove any of these facts. If you dispute them,
>I know where I could find a few, but that might take hours and require me to travel
>for about 600 km. I might be able to google something, but I'm not sure how easy
>that would be and whether I could find something relevant in a language you speak.
>Or I could go to university library (not my home university, however, we deal with technology) and do some digging.
>
>However, my interpretation of this discussion is that you don't even understand
>that these are actually the facts that made me to participate. Not because I think
>you were wrong or Jouni was wrong, but because I don't think you were argueing about same issues.

Well, if you like to correct misunderstandings, you're absolutely free to do so. By all means, please point out what you think were the different issues.

>[Deleting irrelevant parts, about things already discussed above and deemed irrelevant
>there, about things discussed and misunderstood a message or couple ago already.]
>
>> I think I answered clearly what you asked:
>> Forgetting the broken logic in (3)
>
>I think that that broken logic is exactly what shouldn't be forgotten, but this
>is addressed by the great three at the beginning of the message.
>
>> Please show me "the theory" that states what you
>> say. I'm very much interested in seeing the
>> references.
>
>I probably should have corrected you before, but thought you were just being an
>ass and let it slide. I never intentionally referred to any particular theory ("the
>theory") but to the act of discussion taking place on theoretical ground. The "basic
>truths", or some of them, could be said to be theories, but I only brought them
>up to make clear(er) that I'm having a theoretical discussion. These "theories"
>are not meant to be anything new or anything that could be used to refute something
>at such a high level as a processor is.

You know, I decided to check your first post:

"I don't think you have gotten any of the discussion. It is obvious that Jouni has been talking from theoretical standpoint. Sure, you have pulled him in with some idiotic comparisons of real chips, but the fact is that no one has ever made an out-of-order chip by taking and in-order design and slapping out-of-order on it. ..."

Theoretical standpoint usually means something abstracted from reality/details, but sometimes it's simply used to defend a totally bogus argument. The arguments were just that -- bogus. OoOE can't "take 2 extra stages" etc. as asserted. This is not a "theoretical standpoint", but a load of BS.

>> Does his theory take proper account of people
>> whose primary goal is to be misunderstood? :)
>
>I think that was probably studied, but deemed to be lost in the noise :) .
>
>> Unfortunately I'm not aware of practical examples
>> and my own practice doesn't extend that far to
>> know the answer which would be theoretically
>> correct. But since you're the man of theory
>> (apologies, fanboy of theory), you might be able
>> to find it somewhere. :)
>
>As I said above, I'm pretty sure I could find relevant studies, but that would take some effort. For a variation of the theme, there's an empirical test you can try with your friends: http://en.wikipedia.org/wiki/Chinese_whispers
>.
>
>But in all seriousness, the subject is studied a lot and from layman perspective
>it is astonishing just how broken the communication is, and how much of what is said or written gets ignored.

Yes, and one doesn't have to look far: in this very thread you can find numerous examples where communication doesn't seem to work or words like "first" are mistakenly interpreted as "last". :)

>> Err... I don't want anything from you, you're
>> the one who joined the discussion (when it
>> was actually over), so... Did you consider
>> finding another finished thread here and jumping
>> in there as well?
>
>I'm reading RWT once a day, between 6 am and 11 am my local time (now it is 12.33,
>but that's because I'm writing. I was done reading a long time ago). I never response
>to old messages, so the discussion had been over for about 24 hours at most (I don't
>care to check). So I didn't understand that it was over.
>
>I do think that it hadn't been going anywhere for a while and I did hesitate a
>day or two or three before I decided to jump in. I was hoping to stop the stupid
>conversation where neither side really seemed to listen to what the other said.
>Since you were the more aggressive one who did seem to have less glue, I responded
>to you. Note that this is just my interpretation and I'm perfectly willing to admit
>that I may have gotten it completely wrong. I also have to admit that I haven't achieved my goal.

Well, if your goal was *really* to fix misunderstanding, all you had to do was point them out. But instead, you say particular person doesn't get anything and yet you don't point out any misunderstandings. This is what makes me wonder as to what your goal really is.

>> I think it's more up to you to decide what
>> you really want, I don't really care much. :)
>
>Well, it would be great if you would admit that it is possible that you don't always
>understand what other posters mean. That not all idiotic arguments made by others
>are because others are idiots, or are wrong.

Heh, I can tell you that you can find at least one post where one person asked what was meant, but no clarification followed.

>> If anything, I'd like to get an idea of what kind
>> of books or papers (the "Theory") gave you the
>> understanding that you have demonstrated. That
>> would be interesting.
>
>I could say University Physics by Young (and some other people) but to be honest
>I never really read that book, which is why I stated that I haven't studied physics
>in English. So I think I have to go with high school physics books, whose names or writers I don't remember.
>
>If you are referring to circuit level issues, I have only attended to couple of
>basic courses discussing those and don't pretend to have any formal education on those.
>
>What I have done is spent hours talking to students and working out wordings to
>make exercises (both in Finnish and English) as unambiguous as ever. Only to see
>someone completely ignore the context, misunderstand the point and come yelling to us that we are idiots.

Well, it appears it only shows that careful working out the wordings was a waste. :) Someone ignored the context anyway, while others probably wouldn't need the extra work you put in to understand things correctly (or just ask if in doubt).

>-JLarja