| Beitragsrückblick für (die neuesten Beiträge zuerst) |
|
3x0 |
Erstellt: 20:28 am 28. Aug. 2004 |
dammit  
i wish i could hit 2600mhz w/air cooling .
but i must say that that sp97 rock heals the burning wounds with the 92mm zalman over it.
i ll hit h2o cooling in the future.
BTW, anybody using nf7-s, can you recommend me your favorite modified bios?
|
|
smoke81 |
Erstellt: 18:30 am 28. Aug. 2004 |
That's right. The only way could be, using this CPU in a MB with Via or SiS-chipset after modding it to a mobile-CPU. Then you could change multiplier on-the-fly in windows and perhabs rise the FSB with another program (depending on your board and PLL).
But in your case, i see no sense to do this. I think also, that you have already reached the maximum overclocking result, which you can make with this CPU (except compresser-cooling  ).
|
|
3x0 |
Erstellt: 16:14 am 28. Aug. 2004 |
| i see, so there is no way of changingthe multiplier of this cpu ,right pal?
|
|
mastaqz |
Erstellt: 16:09 am 28. Aug. 2004 |
sometimes you get better results, sometimes not.
me can't notice better results in benchmarks
|
|
3x0 |
Erstellt: 16:02 am 28. Aug. 2004 |
thanks for the reply,
i just bought my cpu, and probably it s one of those super-locked cpus you mentioned. currently i m pushing the cpu running on abit nf7-s[v2] to 11x228mhz @ 1.9V (2500Mhz) w/air cooling(thermaltake sp97). also 2x256mb s of Twinmos pc3200 Memory(pretty cool sticks)
i would like to run the cpu at 12 x210Mhz if i could, so i checked out the ocinside guide. but , you say that i can not modify the multiplier of this type of cpu anymore, even with the wire method, this is bad..
i heard that faking the motherboard that cpu is a 200Mhz cpu would let me hit higher FSBs , is that right?
|
|
smoke81 |
Erstellt: 13:45 am 28. Aug. 2004 |
Hello!
Please give us more details, for example: what kind of MB have you? which stepping and rating has your CPU?
First, you get more speed by a higher FSB!
If you have a barton 2500+ (multiplier 11 @ 166 FSB = 1833 Mhz) it would be faster by mult 9 @ 200 FSB = 1800 Mhz! Certainly mult 11 @ 200 FSB is much more faster... That's the same performance of a native Barton 3200+ ...
But there is to say, that AMD has implemented in later produced CPU's (effective from august 2003) a so-called Multi- or Superlock. That means, that you can't change the multiplier of such CPU's anymore.
For overclocking you must rising the FSB. Modern MB's have options for this in the BIOS or at least jumpers to bridge on the circuit board. Unfortunately there also exist boards (mainly cheap boards of complete solutions) that haven't any options for changing FSB.
If you have such a board without possibilities (ch. FSB), then you need the wire method to fake the MB FSB 200. If you have another board then rise the FSB step by step and prove stability by the program "prime95" (torture test).
There is the problem, that only few CPU's (except the lowest Barton 2500+) make the FSB 200 (@ the standard multiplier 11 or greater) at standard VCore (Core Voltage of the CPU, normally 1.65 V) with absolute stability. You must then also rise VCore (Bios or jumpers or pinmod -> see workshop on the left side).
With higher VCore you also get higher temperatures, which should be lower than 60 centigrades or better lower 55 centigrades.
greetings,
Tom
|
|
3x0 |
Erstellt: 13:05 am 28. Aug. 2004 |
hi guys, sorry i cant understand deutsch but i still got a few questions about this site's guides. if you can answer it , id be grateful.
i have a locked barton cpu and i ve checked out the guide at ocworkbench, and i decided to try the `unlock multi via bios' and the '200mhz fsb' solutions for barton. but before i do that id like to ask a few questions.
barton works at 166 mhz default, and the guide represents that to fake the mb for 200mhz, i should add one more wire than the 166mhz config. thats a weird question , i know but if any one ever tried that, he d understand me.
guide also says; by just pinning one wire in the socket, i d unlock the cpu multiplier via bios upto 12.5x. is that really that easy? i mean this is a taboo; unlocking a barton..
and if it d work and i d unlock the cpu's multiplier, (here comes the most important Q) would i be able to hit more speed using a higher multiplier and a lower FSb? if it d not than there s would be no reason for me to wire the socket.
any one ever unlocked a barton using that way?
thanks a lot for reading ..
|
![]() × |
