Hmm, very interesting, so all this time we were fooled by the latency of the i7 by Everest 4, that sucks, i was used to the 34ns!! hehe its ok as long as its normal. I am getting 52ns latency at 1528, cas 9
Printable View
Hmm, very interesting, so all this time we were fooled by the latency of the i7 by Everest 4, that sucks, i was used to the 34ns!! hehe its ok as long as its normal. I am getting 52ns latency at 1528, cas 9
Apparently so.Quote:
Originally Posted by GAR
So one of the theories that the on-die memory controller was giving us much lower latencies appears to be debunked by this...
Comparing bios F3, F4j, F4m, F5g, and F5 (not yet final F5), they all have nearly the same everest score and spi 1m score, within normal variation run to run. Even though slight mem timings differences, they make too little impact to be noticed in variation. Maybe running 32m spi you could see some difference.
Here is F4j versus F5 (not yet final?) showing nearly same everest score/spi 1m with most recent beta build 1658 of everest 5.0. (pic 1 and 2).
However there are still some bugs in it, see pic 3 one weird run using F5 bios and everest 5.0 same most recent build. this bugged run, showing bclk as 242 and mem at 1050 was run right after normal run above on same boot, same bios, just clicked start again and gave crazy readout...so they still have some bugs to deal with.
thanks for the comparisons rge :)
my friend also had a bugged run on his DFI x58 with the new everest 5 saying his bclock was at 300 :rofl:
i thought it quite strange that all the bandwidth and latency readings increased too though, almost as if they were calculated instead of measured :shrug:
I had wondered the same, as it does that on several of my bugged runs, have one 27000 read speed. Seems to read mhz correctly, read multi incorrectly and calculate incorrectly the bclk based on wrong multi. But as to why that would cause bandwidth and latency to reflect the bugged values, instead of actual, I have same thought :shrug:
this was my m8's bugged run
http://www.wright-pc.co.uk/everest/ever.png
that's some sick slices of ram there lol. everest is about at reliable as a Ford Pinto in a demolition derby.Quote:
Originally Posted by Pyr0
is the dual boot problem still going or has anyone been able to get past it. i have tried f5g, previously f5, and f5e also caused that problem any input on what could be done would be greatly appreciated.
thanks for the comparisons rge :up: tryed everest 5 several times to compare since we're running at same speed and timing, i have 3x1 Corsair Dominator PC12800 CL8 TR3X3G1600C8D 7-7-7-21-1, values was about the same as yours everytime. this new version of everest could be reliable after all:p:
http://www.saunders.fr/images/oc7.jpg
i have no more issue with dual boot since i got F5g, id try load optimized default, reboot and enter new settings or you may want to give a try on F5h newest version to date :) wish you luckQuote:
Originally Posted by technut36
more new bios's posted by stasio on tweaktown: http://forums.tweaktown.com/f69/giga...st-bios-28441/
And...Quote:
Originally Posted by stasio
Quote:
Originally Posted by stasio
F6A is also on the gigabyte website
1. Beta BIOS
2. Fix abnormal messages from EasyTune 6 and DES in Vista
:up: Awesome! Running the Feb 12 version of F5 and that actually removed dual booting for me - none of the F5x's removed the dual boot from my machine. Gonna load up the new Feb. 17 version of F5 when I get home.
For whatever reason, I'm reluctant to try out F6a right out of the gate. To whoever gives it a shot: let us know!
EDIT:
Interesting... I've never really cared for EasyTune, so I guess I won't be considering it at all.Quote:
Originally Posted by GAR
I tried the F5 2/17 version (latest) and had no issues with it, performance same, and no double boot problems. Currently running F6a, no problems with it either, works as well as F5 final on mine, not on Vista have windows 7 so cant attest to any bug fixes on those. Performance wise pretty much same.
F6a fixed the LAN problem for me
Am I correct in assuming that F6a for the UD5 extreme is the same revision as F7a for the UD4P?
Hello guys. :) I got my UD5 last week with 6GB set of Corsair Dominator RAM, the problem is that my Vista 64 shows only 4GB installed :(. Everest and Sisoftware show all three sticks. do you know what could be the problem. BTW i have only 500w PSU can it be the problem?, I don't wanna spend more money on something i don't need.
You might want to check if any of your CPU pins are bent.
On another note, although the F5 bios did fix the double boot issue (computer does not have to actually turn off and turn on again) are you guys still experiencing a slightly longer than normal boot time when overclocked? (IE. the small blue LED CPU/BClk speed lights reset like 3 times before actually posting properly when overclocked, but stay solid when not overclocked)Quote:
Originally Posted by PHiZiK
f5g works...x10 on memory not that stable...
going to f6a...will post results later
Good to hear F5 fixes the double post issue.. whew. but if the board still warm resets three times that sucks.. each time it does that my damn RAID card beeps LOUDLY... seems like the further I OC the more it does that warm reset.. I am up to like 5 or 6 beeps on the raid card now. it's starting to get on my nerves.. I can't disable the beep either... I have no idea way Areca thought this loud of a beep would be helpful on each boot?? craziness... I am thinking about soldering the buzzer off the card. :p
What LAN problem? I just started having a strange issue where I need to reset my LAN port to get on the net now after each boot.. kind of annoying.. but only takes a few seconds to fix.. is that what was happening to you also?Quote:
Originally Posted by godsfist
checked the pins nothing looks out of order :shrug:Quote:
Originally Posted by Bluefusion
f6a works great with x10 multiplier
http://i24.photobucket.com/albums/c5/hondacity25/gt.jpg
Anyone else having problems with the Jmicron SATA ports randomly disconnecting external drives? I have my WD 500GB external connected to the GSATA ports and it randomly loses connection with the PC after a while, I need to reboot the PC to see it again. It doesn't matter whether the ports are set as AHCI or IDE, the disconnect issue still happens. I tried updating to a newer driver directly from the JMicron site, but the issue persists. So, are these ports junk, and I need to use the Intel ports, or is there something I'm missing?
using F6A, so far so good.
Can you guys tell me what your experiance with your OC'ed systems have been like when updating the bios?
My i7 940 is clocked @ 3.66Ghz with 1.3v, on the F4 bios it is fine and rock solid from my testing and eceryday use. When I flash to a newer bios, F5(?) or F6a, I load the optimised defaults and everything works and it boots fine. When I try to re enter my previous OC settings, my pc plays up. I'm not getting the "on/off/on" dual boot problem , what happens with my pc is that it boots as normal untill it just gets to the bit were the vista loading bar is meant to pop up, which instead the pc goes back to the "EX58-Extreme" splash page and it gets stuck on that loop. The power does not turn off whilst the pc is doing this. As soon as I flash back to the F4 bios, everything works perfectly. I didn't try and up my voltages to see if I could get windows to load as my question is why is my OC stable with the F4 bios and not the newer Beta's? There is one version of the F5 bios which worked on my pc with the current settings but it would double post if I upped the RAM to 1600Mhz which it is specified to run at.
your not trying to load an old profile right? your are just re applying your settings from memory? Long shot but try disabling the splash screen? Check your boot settings, maybe try having your HDD #1 and disable all the other to make sure there is no conflicts there