Sign in with
Sign up | Sign in
Your question

AMD heads put this in your pipe and smoke it!

Last response: in CPUs
Share
October 11, 2001 10:58:46 PM

Ok we know AMD is ahead in terms of price and work done per MHz and we know Intel is currently ahead in raw MHz. Now can you imagine how awesome it would be if AMD was ahead in terms of raw MHz too!!!! I'm picturing a PR2600+ running 2200MHz.

Now don't tell me MHz doesn't matter because it does. I just proved it.

Remember if you ain't Muslim you ain't Shiite.
October 11, 2001 11:15:53 PM

You proved nothing, except that you have nothing worthwhile to say.

<font color=green>I post so you don't have to!
9/11 - RIP</font color=green>
October 11, 2001 11:28:00 PM

...
Related resources
October 11, 2001 11:45:04 PM

How old are you?

<font color=green>I post so you don't have to!
9/11 - RIP</font color=green>
October 11, 2001 11:49:56 PM

I find this impressive expression of ideas to be very intelligible - wow I'm converted to AMD now!

:cool: <font color=blue> Blowing things up smells aweful... </font color=blue>
October 11, 2001 11:51:15 PM

That will not happen for the Palomino. They pushed their Thunderbird core to the thermal limit at 1.4GHz. Once the Palomino reaches speeds of 1.7GHz it will exceed the power requirements (watts) and heat levels of the 1.4GHz Thunderbird. The Palomino was designed as a temporary solution until they can get their new processor (the hammer line) operational.

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 11, 2001 11:54:47 PM

Quote:
temporary solution until they can get their new processor (the hammer line) operational/


You forget about the Thoroughbred and Barton. They come between Palamino/Morgan and Sledge/Clawhammer.

<font color=green>I post so you don't have to!
9/11 - RIP</font color=green>
October 12, 2001 12:04:34 AM

I do not foresee much of an improvement in these products. AMD is already using many 0.13 micron tricks in their current Palominos. This diminishes the boost in performance that will be seen when they shift from the Palomino. [hypothesis]The Thoroughbred/Barton will likely be only marginally cooler than the Palominos.[/hypothesis]

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 12, 2001 12:34:58 AM

Actually, someone did manage to O/C a T-bird to 2.1GHz. A link was posted here on the board, but I can't be arsed to dig it up.

Kelledin

"/join #hackerz. See the Web. DoS interesting people."
October 12, 2001 12:43:33 AM

Exactly what 0.13u "tricks" are they using? They basically rearranged the transistor layout for more efficiency. They're not even doing SOI yet.

Kelledin

"/join #hackerz. See the Web. DoS interesting people."
October 12, 2001 12:48:00 AM

dunno bout that...

look what happened to the coppermine -> tullatin die shrink.
or the upcoming northwood.
i imagine the voltage reduction/ reduction in heat ouput of new vs old would be similar for all 3 lines (as a ratio)

P.S. can u disclose what voltage your 2.6Ghz is running at raystonn? :D 

Religious wars are 2 groups of people fighting over who has the best imaginary friend.
October 12, 2001 1:54:33 AM

i am so sick of INTEL LOVERs.... whether RESPECTED like RAYSTONN or laughed at like FUGGER....

these guys all have this superiority complex... AMD cant so this.. AMD wont do that ...AMD will not improve that much...

bottom line

as of 10/11/01

AMD ATHLON XP is the KING in performance in X86

What the future holds will reveals itself soon

but right now

INTEL IS NUMBER 2 in performance.
October 12, 2001 2:00:51 AM

As of current benchmarks I'd have to agree with ya.
October 12, 2001 4:29:22 AM

Rayy I must humbly dissagree with you there. If 1.4ghz was the thermal limit of the tbird how come the average overclock for a ayjha core tbird was 1.6ghz aircooled. (athlonoc.com). The palomino can take this figure.(conservativly who knows what process benifits they will get in the next quarter with increased yield and what not) to 2ghz or so. They are releasing throughbred at the end of 1stq next year if I remember corectly, which is 6 months away. I would imagine that they could scale the xp to 2ghz then release the toroughbred at that time to continue a nice linear scale. Then comes hammer.

Please do not make blanket statements.
Quote:
That will not happen for the palomino.

You can not support.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~
October 12, 2001 4:33:45 AM

.13 micron tricks?

Ray I know you work in end testing right, do you even go into a fab, there is no such thing asd a .13 micron trick, you have .13 micron critical dimensions or you do not.

The thorouoghbred chip will most likely get the same heat/voltage benifit which the p4 and p3 gets. There are no tricks in manufacturing. It is a pure scientific reduction, read up on what smaller line widths and critical dimentions do for a process ok.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~
October 12, 2001 4:47:16 AM

Just because someone manages to run their processor at a certain speed does not mean that 1) it is 100% stable and 2) a high yield can be obtained at that speed from the manufacturer. Manufacturer testing is quite a bit more strenuous than those performed by overclockers.

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 12, 2001 4:53:42 AM

Rayy when the AVERAGE, of the overclocks is 1.6ghz, it does in fact mean that the core has NOT reached its thermal limit, and to say it has as if it is FACT is WRONG.

That is all I was pointing out to you rayy, you do exactly the same things you attack others for(or chide them) doing. So stop.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~
October 12, 2001 5:00:28 AM

"when the AVERAGE, of the overclocks is 1.6ghz"

Average for "AYJHA" you mean. And just how many processor are "AYJHA" out of them all? 5%? 10%? Would AMD be happy with a 5-10% yield? I do not think so. When you must look a processor that was made in a certain batch, then it is <b>not</b> the average.

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 12, 2001 5:07:19 AM

Most 1.4ghz tbirds were the ayjha core, do some research before spouting your theories as fact Rayy. ALSO, the other core axia which comprised most of the 1.33's hit 1.5 on AVERAGE. So both your counterpoints are invalid. I accept you know alot more than any of us about the workings of intel processors, but you know nothing more about amd processors than any number of us know. In this case your statement was wrong, if you wish to make "theories" about possible performance of athlon chips, dont label them as fact to take advantage of the respect you have at this forum. I think that is wrong.


~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~
October 12, 2001 5:16:27 AM

Do you have any percentage figures for just how many of these processors were from these specific batches? If most of them were from these batches then why did it even matter from what batch these processors came? Surely it would look like they were all capable of this overclock and this batch likely would never have made the news. Additionally you never addressed the point that processors are much more strenuously tested by the manufacturer than by overclockers. Add to this the fact that AMD must certify all of these processors to run with the default HSF rather than special overclockers' HSFs and you have a problem.

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 12, 2001 5:28:03 AM

I merely was pointing out that
a your claim that the 1.4ghz tbird was the tbirds thermal limit was false.
b you stated it as fact when it was not and you have chided many others including me for the exact same actions.


But I will continue this debate.

The ayjha and axis codes do NOT indicate batch,(batch code actually comes after the stepping code) they indicate process, they are stepping codes. Whenever there is a process change there is a new stepping code to designate this change. To this end, the ayjha process may have been used for 2 weeks or 10,(actually they were used from the release of the 1.4 till a few weeks ago) and there is no way to gauge how many ayjha chips there are in the market. However, i have never heard of or seen a 1.4ghz chip which was not either ayjha(the majority) or axia(minority) both chips are easily overclockable to 1.5ghz+.

As for testing those chips, with time process's mature, and they get cooler and more stable, I have overclocked my friends 1.4 tbird to 1.5 with no problems with the stock hsf. True AMD would have to do a more stringent validation than most overclockers however this does not save YOU from the fact you made a claim which I showed to be false.

The tbird core is not at its thermal limit, if it were it would fail at a much higher rate than it does when you try to overclock, remember the p3 1.13, that chip would not go anywhere, the p3 core would not go over 1.13ghz. THAT was the thermal or otherwise limit of the core.



~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~
October 12, 2001 5:42:02 AM

I will have to disagree with you here. In my opinion, the thunderbird is most definately at its thermal limit with a default HSF and stringent testing standards while still being able to retain a high yield.

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 12, 2001 5:45:00 AM

They act like they are losing sleep over it too. Supposedly the northwood is pushed backed to January now too. The sharpness of the AMD attacks increases whenever performance crown is in danger (or lost) too. As you mentioned, today most people would say XP is the x86 king. In January (or tomorrow), it could very well change again. In fact, I hope they keep changing crowns every few months. $100-$200 for a top notch CPU will be here to stay.
Not everyone wants to or can afford to buy products made by the Intel gods.
October 12, 2001 5:51:14 AM

Which is fine, I cannot assail your oppinion rayy, but it is just that, your oppinion. You stated that it was a fact the 1.4ghz tbird was at its thermal limit, which is clearly untrue. You can believe whatever you wish.

However some points to ponder.
A: standard heatsinks change all the time, whos to say that amd cant start packaging swifttech heatsinks with its 1.6ghz tbirds(theoretical mind you). So that reasoning is flawwed.
B: there is yield (working die) and YIELD(the speed at which the processor can run reliably). They shouldnt be called the same but they are. If AMD makes a die which cant cut it at 1.6ghz they can just sell it at 1.4 no big deal. The Thunderbird core is not thermally limited from performing at 1.6 or even 1.7ghz, to claim so as a fact is wrong.

~Matisaro~
"The Cash Left In My Pocket,The BEST Benchmark"
~Tbird1.3@1.5~
October 12, 2001 5:54:06 AM

"You stated that it was a fact the 1.4ghz tbird was at its thermal limit"

Would that be because it is? ;) 


Anyway, we are just going to keep going back and forth here. This is losing its novelty. How about we head over to the retro thread and dance to some 80s? *grabs his dancing shoes*

-Raystonn


= The views stated herein are my personal views, and not necessarily the views of my employer. =
October 12, 2001 8:02:26 PM

He'll claim to be 32 and working for some computer copmany, just like fugger. They all follow each other's leads.

<font color=red>umtqyvutkhgjvqffpcvpgrgtfocpgnnchvtcwqjvgepgjyoqthgtqhgtgjvtgdogogt....74pqkvcngxgt</font color=red>
!