I found one under my coleagues desk this afternoon. Test reports soon to folow
Printable View
I found one under my coleagues desk this afternoon. Test reports soon to folow
Quote:
Originally posted by Praetereo
I have not seen anyone posting any OCed results for this board yet, and I wanted to post my friend's results. Hope no one minds :)
This is my friend Aceman's Sandra memory scores with one stick of 512MB XMS3500 @ 2.7v.
http://webpages.charter.net/praetereo/233_27v.jpg
i hope this is with relaxed timings
My thought exactly!Quote:
Originally posted by spaceboy
i hope this is with relaxed timings
A better pic (with timings) would be nice.. also more info.. 234 is nice, but sucks if this is fastest timings :D
For that high speed run I posted the screenshot of, he said :
"It ran at CAS2 5-2-2 almost all the way up. Someware towards the end I had to raise them to CAS2 6-3-3 since I couldn't throw any more voltage at it." He also says he does not see an option inthe bios to adjust the command rate, so he thinks it is at 2T. I will ask him to check with software.
As far as what kind of stability he has at those speeds, I don't know. But he says 200MHz is completely stable.
Here is the info my friend Aceman gave me about his drive setup:
"Ok, I had three hard drives hooked up and my DVD burner. My main WD was primary master and the dvd burner was secondary master. My two IBMs were hooked up via two converters to the SATA controller in RAID-0 array. I've been running at 200mHz FSB ever since that test and haven't experienced a single lockup or crash of any sort. I even kicked my CPU up to 2gHz for several hours today. That 10x200 config allowed me to reencode a DVD at 2x realtime. I did two passes on the encode and then even cut the DVD including a simulation while still overclocked (kinda forgot I had overclocked it for the encode) So, based on the past few days, I'm pretty much planning on running 200fsb permanently from now on."
His CPU is an XP 1.8GHz.
Also, on the information your were looking for, RacerX, his board is still installed in his case, and he can only see about 75% of the back of it through the frame, and it was not located there. He does not have time to uninstall everything and pull the board out until this Friday, so I should be able to post that information Friday night.
the epox bios now allows upto 250mhz in 1mhz increments...
Gonna have some fun with that later.....
hmm can anyone post me how the 12 and 5v rails scale with a higher voltage (cpu 2.2 / ram 2.9) ?
Could you get your friend to tell you what settings he set in Soft Menu. Such as which divider he was using and all of his voltages. I am testing this board right now and so far it bites.Quote:
Originally posted by Praetereo
I have not seen anyone posting any OCed results for this board yet, and I wanted to post my friend's results. Hope no one minds :)
This is my friend Aceman's Sandra memory scores with one stick of 512MB XMS3500 @ 2.7v.
http://webpages.charter.net/praetereo/233_27v.jpg
I Thought the nforce2 didn't have a pci divider. Am i mistaking this with something else, some other bit of information maybe?
Correct there is no PCI divider. I was speaking of the CPU FSB/DRAM. (By SPD, 3/3, 3/4, 3/5, ...)
It's confirmed by ABIT that the PCI is LOCKED for everyones information:)
Randi:D
Guess I will go ahead and mod pin 7 on Vcore and 6 on vmem tonight to see if i can get this board to a respectable OC before I revert back to the KD. If anyone smells smoke tonight it probably my NF7-S.:banana:
Icee, does Abit know about the 188 FSB issue?
I've been waiting so long for the real NF7-S, and if it cant pass 188FSB using its very own IDE controller, I will be less than happy.
I don't think it's a PCI issue as far as getting over 188. My board won't boot over 188 even with all IDE devices disconnected. CPU and memory (and hard drive for that matter) has done 206 on a KD7, so I don't think they are the culprit.
I've done the pin 7 mod, and it works. I'm able to get to around 2.0v. Going to try the pin 10 mod just to see how this new controller behaves.
Have you tried the Vmem mod?Quote:
Originally posted by sharpbw
I've done the pin 7 mod, and it works. I'm able to get to around 2.0v. Going to try the pin 10 mod just to see how this new controller behaves.
Once I disconnected the IDE drive I was able to boot up as high as 230 with my TwinMos. I find that odd you can't even with everything disconnected. Make sure you have CPU Interface enabled.Quote:
Originally posted by sharpbw
I don't think it's a PCI issue as far as getting over 188. My board won't boot over 188 even with all IDE devices disconnected. CPU and memory (and hard drive for that matter) has done 206 on a KD7, so I don't think they are the culprit.
I've done the pin 7 mod, and it works. I'm able to get to around 2.0v. Going to try the pin 10 mod just to see how this new controller behaves.
What kind of ram are you running?
Hallowed, I pm'ed ya about whats going on with these boards. I have confirmed with Abit there is a issue within the board. They are not able to get passed 188 either. This has nothing to do with the voltages. I can take my TwinMos at cas2,2,2,5 dual channel at 2.6v. at 188.
I have to admit for 188fsb this board kicks butt. I have tried running my memory around 240 but not really much to bragg about. I want that 200fsb dual channel maxed out and possibly higher. I think to go beyond the 200fsb dual channel Abit will need to increase the voltages or we do our own mod.
I'm sending my board back on Monday. If anyone here as the board might want to give a promise or highpoint controller a try. The pci is locked so that just might be the ticket.
Will be doing just that tomorrow RacerX...my Promise Ultra133 TX2 is awaiting:)
Hallowed:)
RacerX is correct that they are aware of it. I haven't had my hands on one yet to evaluate and give feedback to them as the next shipments of the "revised" board will not be here until the end of the month, but I sold some of my hardware and ponied up for a first release version to see what I can conger out of it. I'll be reporting back here this weekend.
Marty....let me know you're results of the Vmods...either here or email or pm....I'll want to do them before installing the board if they BOTH are successful. As you know my 8RDA+ Vmem attempt destroyed the board:(
Randi:D
Your the man Randi. I just killed my NF7. I was trying to see how high I can take my TwinMos in 5:6 mode. I was able to boot up at 480 that rebooted and tried 485 than it froze in bios. Couldn't bring the baby back to life. Unplugged everything, removed battery, tried different cpu, ram, video card, and still nothing.
Oh well atleast now I don't have to pay a restocking fee.:D
Sorry to hear of your loss my friend:( I know this feeling to well myself lately.
Listen up guys....running these boards async is fruitless. We only have certain bandwidth bus to work with which is double pumped ie; DDR. Running the ram faster does not benefit an AMD...and latencies are not the reason. I ran my ram at 233Mhz on the 8RDA+ with a 200Mhz FSB. 466Mhz bandwidth pumping into 400Mhz bandwidth capable bus...it's like a guy with a 46" waist trying to funnel his fat ass into a pair of 36" pants:D
Intel boards can pull this off because their FSB is quad pumped. Look at JC's IT7....500 Mhz bandwidth DDR ram(250mhz) feeding a 180Mhz FSB x 4...or 720Mhz FSB bandwidth. Thats why it works for Intel and not AMD.
Randi:D
I can only agree with your statement. My benchmarks were little to be desired.
Anyway how did the 8RDA+ work out for ya Randi. I have one arriving on Monday. I'm already feeling regrets about ordering it.
Yes, the question is can the next NF7-S go above 188, and if not, is it a BIOS problem or manufacturing flaw.
People have Epoxes running @ 230+, no reason why Abit shouldn't.
Then again, the Epox KT400 board can bench 3d @ 240 due to some oddball divider @ that speed.
The 8RDA+ is a good board RacerX. I'd still be playing with it if I hadn't had a failed VMem mod attempt. Purely my fault indeed.
Randi:D
you're right but whne going with fsb higher than ram frequency you might have that "intel effect" but you're not indeed... it's there that memory controller (and probably latency) gets all worst...Quote:
Originally posted by MrIcee
Sorry to hear of your loss my friend:( I know this feeling to well myself lately.
Listen up guys....running these boards async is fruitless. We only have certain bandwidth bus to work with which is double pumped ie; DDR. Running the ram faster does not benefit an AMD...and latencies are not the reason. I ran my ram at 233Mhz on the 8RDA+ with a 200Mhz FSB. 466Mhz bandwidth pumping into 400Mhz bandwidth capable bus...it's like a guy with a 46" waist trying to funnel his fat ass into a pair of 36" pants:D
Intel boards can pull this off because their FSB is quad pumped. Look at JC's IT7....500 Mhz bandwidth DDR ram(250mhz) feeding a 180Mhz FSB x 4...or 720Mhz FSB bandwidth. Thats why it works for Intel and not AMD.
Randi:D
Quote:
Originally posted by Hallowed
Then again, the Epox KT400 board can bench 3d @ 240 due to some oddball divider @ that speed.
tell me more 'bout it :D
Let me dredge some photos.
A friend of ours at TNR (UmeKami10) managed to get to 240 FSB. He said anywhere between 204 and 238 or so and artifacting was God-awful. After booting at 240, there was no screen corruption, drive corruption, anything. So he ran a 3dMark.
It scored less than his run @ 200. Odd §§§§, I know, but this makes the second Epox with a hidden divider.
Let me find the screenshots.
Direct cut+paste for the curious. This is about the 8K9A2+ or whatever the Pox KT400 is. Keep in mind my bud kami can overclock like hell, but is a total newb at 3dMark :D
Quote:
I am not stopping to be amazed of this board.
Here is story. I install Geforce 4 ti 4200 and a IBM hard drive ATA 100 /30 Gig in order to se how high this board can manage to run 3D mark. I am not gamer and in fact I don’t like 3D mark and I don’t see any meaning of that but it appear to be very popular …. I got a few mail and some people who asked me to run 3D mark in high FSB ……
I have worked on this project nearly 8 hours. Hear is my finding:
1- It is almost impossible to run 3D mark from 207 to 239fsb. From 236 to 239 nothing Then suddenly from 239fsb is possible and work very nice.
2- File corruption on this IBM drive start 210fsb to 236 fsb. From 239 fsb it work great.
I paste 2 screenshot. The first one on 200X8.5 =1700 Mhz and second 239X8.5 =2034 Mhz.
Bios setting are default except AGP 2X and Memory setting on cas2 and 4 bank, sound and networking disabled. Everything else is default.
As you see the result of 239fsb is lower than 200 . And as I said there is no problem to run HD drive and Geforce on 239 fsb. How should we interpretate this results.
Is it possible that we got 1/7 divider that automatically start working when PCI buss increases 239FSB ? I put my 5 cent yes there is.
Why the result on 239 fsb is lower than 200 fsb? Well it must be some bug in bios . We should try to find thr register setting in Wpcredit then this board might outperformed the other AMD board in 3D mark?
XP 1600+
MSi Geforce 4 ti 4200 OV 310/610
256 mb TwinMos cas2 /4 bank
IBM ATA 100/30 GIG
Sound and lan disabled in bios. Agp run on 2X. Everything else defaults.
http://hem.spray.se/umeimak10/epoxpr...0fsb3dmark.jpg
http://hem.spray.se/umeimak10/epoxpr...9fsb3dmark.jpg