Printable View
No I was too busy those couple of days......I'll check it on Saturday coz I wan't to find out the VCore Mod of the ABIT mobo and prepare it for LN2 benching, so I'll find some time to do it before plastic spraying it.....;)Quote:
Originally Posted by cirthix
awesome :D
updates? (hopefully ;))
hipro is very busy at this moment with the aw9d max, far more people are waiting on mods for that then there are looking for this.
Not only that, if hipro can do this, odds are we still wont be able too, so its not as important as it is for him to break some WR's on the abit.
My point, just be patient, the master can only do so much! :p:
well it seems that the new kentsfields are displaying fsb walls at 350fsb and often lower :(
what we need here is a proof of concept mod to prove that vcca is indeed the culprit.
then we can twist the manufacturers arms properly for them to include variable vcca in the bios.
Vfsb should be kept at 1.55Volts max(as for Intel - speks -)......VCCA pin STARTS from 1.5Volts normal with maxed at 1.8Volts......Nowadays mobos which feed the Vfsb WITH the VCCA together, start from 1.2Volts which is the default for the Vfsb.....VCCA is getting TOO LOW Voltage in such cases.....Raising the Vfsb/VCCA from 1.2Volts to 1.55Volts you could gain from 30 to 65MHz fsb......Though I'll see what could I do about the VCCA mod now that I have some spare time this week.......:)
is this the case for D975XBX Bad axe as well ?Quote:
Originally Posted by hipro5
ALL of the mobos are EXACTLY the same......It must be a reference CPU circuit from Intel which ALL of them follow by the letter........;)Quote:
Originally Posted by eva2000
If intel artificially limited the FSB limit on their cpu's and chipsets, how come some are hitting higher speeds than others? It's all just a luck of draw if you ask me
Quote:
Originally Posted by _Eduard_
They didnt limit anything, what hipro is saying is that the way intel mobos are designed its impossible to give more volts to the VCCA without overvolting other parts of the cpu...
but if we isolate the vcca on the cpu from teh vcca on the board by removing the pin or coating it with an insulator such as nail polish, it would be possible to feed the processor a vcca from an external source.
I've been reading this topic... it's been a few days since no news... did hipro do the trick?
you guys are forgetting the chipset here, it does limit cpu overclocking as there i s a direct link for NBCC and the CPU CC and we all know the NBCC will max out first.
I agree VCCA needs to be seperated and will help but to dismiss the chipset is plain wrong. Chipset yield is also important along with the internal latency settings.
Quote:
Originally Posted by Tony
Yup, but the point of this is for chips that arent limited by the chipset, like my old 6300, couldnt pass 520fsb no matter what, even though it had way more in it.
my l629b405 on a p5b-deluxe does 535fsb.so no wall there i dont think.
http://loader.sc0rian.com/users/publ...c2d100n197.jpg
my L628A E6400 has a strong and stupid FSB wall at only 455mhz... my mobo can do 455mhz with all voltages in minimum.
On an EXACT same mobo have been tested three different CPU's.....NONE of them did the same fsb OR at least close to each other.......FSB depents a. on the chipset and b. on the CPU......If we have a chipset that can do 500MHz fsb and we put on it a CPU that can do only 390MHz, we have nothing......So companies must consider of puting an external VCCA circuit for their mobos coz now with the Kentsfields coming around the corner, we will face a big problem......
One might have a GOOD mobo that does 600MHz fsb and then put in it a Kentsfield and gets 370MHz......What he will think of?....That his mobo isn't capable?.....He DID 600MHz.......Now why not?.....;)
P.S. I haven't experament yet with the VCCA mod coz my rig is in it's "benching days" and I don't want to touch anything for now.....I'll do it next week...... :)
Quote:
Originally Posted by hipro5
one week, i'll be :stick: ing you then hehehe.
another idea: burn a trace. if the vcca and vcc are indeed shorted together with a trace, it would be easy to remove the cpu, solder a few leads to a couple vcc pins, solder a lead to the vcca pin, then push a good bit of current through it for a few seconds to burn the trace connecting vcca and vcc. its' a bit risky, but it would isolate the vcca pin from vcc so that we can solder a wire to the back of the board to a power source (which could be as simple as a voltage divider from the 3.3v line with a pot and fixed resistor, maybe an added cap to help filter some noise).
it's risky, but would work better than insulating or breaking a pin from the socket.
Do you think vcca will help, remember Yonahs? Increased vcca gave 5-10fsb at most
From what I have seen so far and tested myself subzero cooling helps to increase fsb wall barrier 30fsb +
it will likely help kentsifelds a lot more than conroes, but I do beleive that it will help a little. I dont think i'll be doing it to my chip, but if it helps, i will consider it.Quote:
Originally Posted by kiwi
also keep in mind that oc'd conroes are operating at speeds exceeding double what the best yonahs did.
outch!!!! i gues it's just luck thenQuote:
Originally Posted by hipro5
anyways, can someone link me to the vNB mod?? is it easy to do?
i'm trying a 6600 with very low fsb wall near 380mhz, i have test this mb with a 6300 for 1 month RS at 470mhz and fsb wall near 490mhz.
Heeeeelllo Hipro ;)
We need more fsb :D
I can partially confirm that FSB wall has been hacked :D A guy of this forum knows details but he won't share them yet :)
So fsb wall is possible to hack :)