Super Bypass enabler for AMD Northbridge 750 chipset (K7M)
Kevin Peterson
a feature that if supported can be enabled and thus removing alot of memory
latencies handled by the chipset.
The problem is that i haven't heard anyone mentioning about this new feature
but if someone out there is a reaaally good programmer he could right now in
this very moment just write a patch that goes into the PCI registres and
detects the feature and then turn it on..
There is no drawback against this feature its is only lacked just because of
earlier revisions of the AMD Nothbridge 750 had a problem with that
function but there are alot of us outhere that doesn't know if we have the
new or the old revision of the Northbridge. If we have a newer we could as
well turn it on giving it a increase of 25% on the memorylatencies.
Check out the article at Tom's Hardware Guid if you're interested in it and
please give a reply if somone knows anything about this..
Kind Regards Kevin Peterson
DeadFish
> but if someone out there is a reaaally good programmer he could right now
in
> this very moment just write a patch that goes into the PCI registres and
> detects the feature and then turn it on..
Are there any documents describing the register-level interface of Super
Bypass?
--
Remove ANTISPAM to reply via e-mail
Kevin Peterson
can see anything.
You can read the article on Tom's Hardware to gain more information.
Regards Kevin Peterson
Slash
scribbled:
>As far as i know there isn't for now. Check around on the AMD website if you
>can see anything.
>
>You can read the article on Tom's Hardware to gain more information.
>
>Regards Kevin Peterson
One's already out. Check www.aceshardware.com for the link.
-Slash
"In the beginning, there was nothing, which exploded."
- Terry Pratchett
Kevin Peterson
Bypass function or not, clearly the ID program says revision A of the 750
Northbridge. You have to have the C revision to actually got it working.
Doh! :-/
Wonder if there really are some K7M mobos out there with revision C of the
Northbridge??
Mr Pants
Yummy.
It's a white boxer, bought in October in the UK
--
=====================================================
Reply to: ch...@NO.SPAM.theskirrid.demon.co.uk
'shedding skin succumb defeat
this machine is obsolete'
The Pentium 3 song from Nine Inch Nails
=====================================================
Mr Pants
Groovy.
It's a white boxer, bought iin the UK in October.
DeadFish
> Thank you for the input. Now i don't have to wonder if i can have the
Super
> Bypass function or not, clearly the ID program says revision A of the 750
> Northbridge. You have to have the C revision to actually got it working.
I tried it on mine...said my chipset was revision P. Sounds a bit odd.
DeadFish
[Super Bypass register description]
> As far as i know there isn't for now. Check around on the AMD website if
you
> can see anything.
First place I tried: I couldn't find the nescessary info :o(
You'd think AMD would give us the tool to activate s/bypass!
Julian Spencer
program available which apparently this time indentifies it correctly.
Previously the old version said I had a Revision A, now it says I have a
Revision C, Stepping 4 and Bypass is disabled.
http://www.amdzone.com/files/chipid.exe
--
Julian Spencer
jspe...@chariot.net.au
"Kevin Peterson" <l...@telia.com> wrote in message
news:4Ntb4.2205$lP4....@newsc.telia.net...
> Thank you for the input. Now i don't have to wonder if i can have the
Super
> Bypass function or not, clearly the ID program says revision A of the 750
> Northbridge. You have to have the C revision to actually got it working.
Kevin Peterson
Julian Spencer <jspe...@chariot.net.au> skrev i
diskussionsgruppsmeddelandet:386ec...@news.chariot.net.au...
> Did you use the Chipset ID program? If so there is a new version of the
> program available which apparently this time indentifies it correctly.
> Previously the old version said I had a Revision A, now it says I have a
> Revision C, Stepping 4 and Bypass is disabled.
>
> http://www.amdzone.com/files/chipid.exe
Well look at that, now it says that my mobo has the revision C chipset too
and thus it is disabled.. DOH! I reaally want someone to write a nice tool
to switch it on.. BTW does that program enable it?? I don't have a clue..
Regards Kevin Peterson
Sample output
<---Snip--->
Chipset Identification Utility v0.21
(c) 2000 Tamas Fejerpataky
Chipset vendor.......: AMD
Device...............: AMD-751 "Irongate" AGPSet
Revision.............: C, Stepping 4
-----------------------------------
Bypass Status........: Disabled!
<---End Snip--->
Julian Spencer
adding the feature to enable/disable Super Bypass if it works. Also I'm not
sure how accurate this program is. From everyone I've heard from who has
used this program on various web boards, everyone has had the same
experience of it being reported as Revision A with the old program, then
Revision C with the new program. I'm still yet to see someone have it report
Revision A with both versions or even a Revision B if there is one. If there
is a second utility which can do the job I'd like to know so I can be sure.
--
Julian Spencer
jspe...@chariot.net.au
"Kevin Peterson" <l...@telia.com> wrote in message
news:4fAb4.206$_z9.18...@newsa.telia.net...
Julian Spencer
allows you to enable/disable Super Bypass. Still reports Revision C,
Stepping 4. If you downloaded it before from this location you may have
trouble getting this version as it may be still in your ISP's cache.
http://www.amdzone.com/files/chipid.exe
---
Use the /B+ switch to turn on bypass, and /B- to turn it off. There is also
a
/? for help. You can include this in the autoexec.bat file (theoretically),
like so:
chipid /b+
It may even work from within a Windows DOS box (wouldn't that be great for
benchmarking? :) but I'm not sure. To be safe, try it in raw DOS first.
--
Julian Spencer
jspe...@chariot.net.au
"Julian Spencer" <jspe...@chariot.net.au> wrote in message
news:386ed...@news.chariot.net.au...
Ken Neal
run CHIPID?
I have a K7M using 1209 bios.
Here are my results:
C:\>chipid
Chipset Identification Utility v0.3
(c) 2000 Tamas Fejerpataky
Chipset vendor.............: Unknown
Device.....................: Unknown
Revision...................: P, Stepping 16
"Kevin Peterson" <l...@telia.com> wrote in message
Jeff
Revision A with the first version and then Revision C, stepping 4 with
the new version. By the way, the newest version which I used to try to
enable Super Bypass didn't work on my K7M. It just hung and beeped.
Jeff
remove maker of spam to email
Julian Spencer wrote:
>
> The program just identifies that's all. Perhaps a BIOS update will come
> adding the feature to enable/disable Super Bypass if it works. Also I'm not
> sure how accurate this program is. From everyone I've heard from who has
> used this program on various web boards, everyone has had the same
> experience of it being reported as Revision A with the old program, then
> Revision C with the new program. I'm still yet to see someone have it report
> Revision A with both versions or even a Revision B if there is one. If there
> is a second utility which can do the job I'd like to know so I can be sure.
>
> --
> Julian Spencer
> jspe...@chariot.net.au
>
> "Kevin Peterson" <l...@telia.com> wrote in message
> news:4fAb4.206$_z9.18...@newsa.telia.net...
> >
> > Julian Spencer <jspe...@chariot.net.au> skrev i
> > diskussionsgruppsmeddelandet:386ec...@news.chariot.net.au...
> > > Did you use the Chipset ID program? If so there is a new version of the
> > > program available which apparently this time indentifies it correctly.
> > > Previously the old version said I had a Revision A, now it says I have a
> > > Revision C, Stepping 4 and Bypass is disabled.
> > >
> > > http://www.amdzone.com/files/chipid.exe
> >
> > Well look at that, now it says that my mobo has the revision C chipset too
> > and thus it is disabled.. DOH! I reaally want someone to write a nice tool
> > to switch it on.. BTW does that program enable it?? I don't have a clue..
> >
> > Regards Kevin Peterson
> >
> > Sample output
> >
> > <---Snip--->
> > Chipset Identification Utility v0.21
> > (c) 2000 Tamas Fejerpataky
> >
Slash
<kn...@tampabay.rr.com.zzz> scribbled:
>Can anyone provide info about why the "unknown" is being displayed when I
>run CHIPID?
>I have a K7M using 1209 bios.
>
>Here are my results:
>C:\>chipid
>Chipset Identification Utility v0.3
>(c) 2000 Tamas Fejerpataky
>
>Chipset vendor.............: Unknown
>Device.....................: Unknown
>Revision...................: P, Stepping 16
I'm guessing you're running it under Windows 2000? I get the same
message when doing so on my P3B-F/Celeron @550 using W2k build 2195.
You'd need to run it from a pure DOS box or from Windows 98.
Kevin Peterson
> Revision A with the first version and then Revision C, stepping 4 with
> the new version. By the way, the newest version which I used to try to
> enable Super Bypass didn't work on my K7M. It just hung and beeped.
>
> Jeff
Well then Jeff, we actually have the same problem then.
I did my 3 cents of effort in this case too because when i tried it it just
hang and i got a BSOD. So much for that utility i thought.
But i did a clean install of Windows 98 SE and tried to figure out why it
hung just of the curiosity of it.
This is my experience.
If you do a clean install and put the chipid /B+ option in autoexec to turn
it on under the installation it worked as smooth a baby skin.
When you restart and it will find all the plug'n'play devices in the system
i got a BSOD when it tried to detect the USB controller. Why it hung i don't
know but when i restarted the computer continued from the position where it
hung. BTW USB Controller is disabled in Windows when it was ready. "The
device stopped responding" blaha. you know the rest.
When the setup was complete i began with first of all testing if it did some
improvement and ofcourse it did. Sisoft Sandra MEM reported a increase in
both the CPU mem throughput and FPU mem throughput. CPU from 240 - 265 and
FPU from 199 - 221.
So there you have it a slight improvement on the mem throughput speed.
After that i installed the AMD miniportdriver and restarted. It found the
AGP port and the Cpu 2 PCI hostbridge.. etc..
but when i installed my graphicscard it just went away. I got BSOD's and
lockups on my way in to Windows..
So.. atleast when having a Riva TNT based graphicscard it crashed during
Windows boot.
I tried all versions of Nvidia drivers from 2.08,3.53,3.56,3.62 and finally
3.65 and none of them worked.
If someone get it to work notify how you did it.. because as said earlier
when i installed the drivers to the graphics card it all crashed othervise
the Super Bypass feature worked flawlessly..
My 3 cents :-) Regards Kevin Peterson
Jeff
bypass was broken. This might explain all the problems we've been
having.
Jeff
Kevin Peterson
Jeff <fr...@hotmail.hormel.com> skrev i
diskussionsgruppsmeddelandet:386F8A56...@hotmail.hormel.com...
> If you read tom's article, it says in the older chipset revisions, super
> bypass was broken. This might explain all the problems we've been
> having.
>
> Jeff
Yes!
I know but sources also say that revision C is the revision where Super
Bypass works on, and i have Revision C and it worked!
But not with a graphics card installed. :-/
We can only pray for BIOS update where there is an option to turn it on
instead for an dos utility.
Regards Kevin Peterson
Julian Spencer
chipsets. Most of the people I've heard from have reported Revision C,
Stepping 4 chipsets (including myself) and it hasn't worked. Stepping 4 must
be pretty close to having it working as you've had some success with it but
it's probably not enough to have a system running stable with it. Damn so
close.
--
Julian Spencer
jspe...@chariot.net.au
Kevin Peterson <l...@telia.com> wrote in message
news:sTNb4.260$_z9.18...@newsa.telia.net...