even with later bioses????
Printable View
yes!
i'll all my dominators lying around and some other sticks
if they all top out there i'm moving onto other things
we are getting DFI P35 boards in australia end of this month.......i think i'll give those a shot too
yea, In re to one of your recent posts Ace, Didn't think about it until now but, the performance gain I got with the b4 bios was the first time I flashed using your procedure, b3 was flashed using abits flash menu. that is most likely where I got the performance gain as others have not seen a difference. Either way, Thanks Ace for the good info.
I think the tigher you set your timings in bios, the lower the board sets the performance level... Maybe running memset and adjusting the PL will help or what I do is just crank the voltages hehe.... I think with my Ballistix pc8500s Ive hit ~640fsb cas5 and 620ish on cas4 with default subs.... This one with my ballistix 5300s, it doesnt clock as high as the 8500s but it can handle tight subs:D
http://ocidb.com/albums/userpics/10062/normal_3591.JPG
cas5 low volts... tops out at 620fsb though no matter what.....
http://ocidb.com/albums/userpics/100...l_600stock.jpg
570fsb... cpu limited cos on air I have a fsb wall at around 552fsb....
http://ocidb.com/albums/userpics/10062/570fsb.PNG
Everything was done on original bios.... havent flashed to the new ones yet....
No its the faster you set your ram the lower the performance level goes. I can run 600mhz or higher with my ballistics and the performance level changes with speed.
Like 500mhz its around 7, 550mhz its around 6 and 600mhz I usually get 5.
Timings dont seem to make a difference with my setup.
Ive only tested them with those ballistix.... I think this board just sets the PL level lower than the Asus or Gigabyte thats why its harder to hit high mhz... Ive seen P5Ks with PL of 9 or even higher whereas with my board I cant remember even seeing PL=8....
It does take alot of vdimm to hit 600+fsb cas4 though and I think thats where the ballistix 5300s have a big advantage over the Dominators.... Its alot easier to beat up cheap rams:rofl:
i've got a "few" kits of PC10000 dominators and the best kit can do 600MHz 4-4-4-4 with tight subitmings at only 2.4v man :( (I'm talking 32M stable)
i tried changing the value in windows with memset and it didn't seem to impact it at all....but i've only had a couple of hours of play so far
hitting decent clocks should not be hard though.....so i'll try and retest again./....but my confidence is a little low....
now i'm not sure if it's worth even setting it up for phase :(
It's MemSet 3.3 beta2 by the way..
3.3 final has performance level option for me :confused:
Mine actually sits pretty flat... First thing I did when I got the board was take the heatpipe off, do a lil bending replace the rubber stuff with stg1 and replaced all push pins with screws.... did the same thing with my Commando as well...
http://ocidb.com/albums/userpics/10062/IP35b.JPG
Add a fan on top and it works pretty good....:D
http://ocidb.com/albums/userpics/10062/fans3.JPG
1.72v... Here's the whole screen....
http://ocidb.com/albums/userpics/10062/normal_5130.PNG
Ok
And, what your max fréquence FSB for 1.25Vmch (strap 1333 & 1066) plz ??
srry, with memset, PL5 is better than PL7? or the other way around.
I honestly dont have a clue.... Its my benching rig so Im always testing for something else and most of my voltages are probably too much....
Its L632B172... Good clocker for air and water but not really all that great subzero cos the fsb wall is pretty low....
Lower is better
That's with beta bios b04, the board feels very strong, and think i'll be reaching a cpu wall soon.., next will be FSB 493:up:
http://img291.imageshack.us/img291/2...2forum2ux2.jpg
Just acquired a pair of 512 x 2 Kingston DDR2 800 ValueRAM part number KVR800D2N5K2/1G. Managed to run them at 604MHz with 2.1Vdimm and 5-5-5-15-2T. There's a logo on the chip with six small squares in a 2 x 3 pattern and two lines surrounding this pattern. Anyone familiar with this logo? Sorry but no digital cam.
____
***
***
____
Those "N5" modules generally overclock well. Kits without N5 are normally binned at 430MHz.
it is very obvious that abit boards work better with kingston modules.
what processor do you have?
E4300 (L2 stepping with 100C Tjunction) at 3.56GHz with 1.505Vcore. Divider set at 1:1.50, 2.1Vdimm, 5-5-4-4-9-2T. Room temperature is 89F. Will run Orthos for 1 hour during the winter. No need to stress my CPU with high Vcore and temperature right now.
VTT, VNB, VSB at default.
Got very strange problems when i try to open Uguru.
Installed my IP35 Pro yesterday and a new HD 2900XT. Today i updated bios to b04. Did a run in 3dmark2001 and got 47900p with cpu and gpu at stock.
Using the latest ATI driver to. What can be my problem u think ?
My system:
Abit IP35 Pro
E6600
Asus HD 2900XT
Xtreem PC6400 2gig
Xp Pro
Hope someone can help me
Look at the pic and u know what i mean ^^
http://img166.imageshack.us/img166/2...angewn1.th.jpg
That looks like some sort of video corruption to me, most likely something to do with your new video card. I've seen that when temps are too high with the video card, check the cooling on it maybe?
are you running in "dual" monitor scenario...my son had something similar and it was due to the driver running in the dual monitor mode...check that first...if that is not the problem, then uninstall the driver, reboot, and reinstall it again.
Running with only 1 monitor. And the temps for gpu are around 60-65 when just surfing. It seems only got problem when i open Uguru. Everything else works fine, can play without atrifacts. But´s hard for me to use uguru when i can´t see everything. When i open firefox the black box on the screen disappears.
Hmm really strange....
You should also uninstall uguru and reinstall it, preferrably with the latest version which can be found in this thread.
no, no, no!:) ...he was running only one monitor but the driver through the control panel was set up to run in a dual monitor mode...as soon as he changed it to a single monitor mode, his video corruption went away...check your video driver configurations through the control panel.
Has anyone got anything good to say about the IP-35 Vanilla? Nearly a £40 difference between it an the Pro. Would be nice if it OC'd well :)
i would go with the Pro because you have Uguru...that little prog is a jewel in my estimation...you can change any voltage, pcie freq or FSB from windows...it makes it so much easier on you when you are trying to find the right combination of voltages and FSB setting for your maximum overclock.
I may bump up Vdimm if I know the mfr of these chips. Very strange logo. RAM is rated at 1.8V.
Just discovered a possible RAM anomaly with my Abit IP35-E/E4300 CPU combo. 1st pair of Kingston DDR2 800 (chips on both sides, 1GB x 2 with SPD 5-5-5-18/1.8V) is set at 5-4-4-9-2T in BIOS. Memory clock speed is 480MHz (overclocked CPU with 1:1.25 memory divider). Memtest 86+ V1.70 shows bandwidth of 5628MB/s.
Swapped these RAMs with 2nd pair of Kingston DDR2 800 (chips on one side, 512MB x 2 with SPD 5-5-5-18/1.8V). Same memory settings as the previous pair. Memtest 86+ V1.70 shows bandwidth of 5197MB/s.
Not sure why the speed of the 2nd pair (512 x 2) is lower than the first pair (1GB x 2) when all other parameters in BIOS are the same. Both pair are stable up to a minimum of 576MHz.
Memtest reports a bandwidth of 4358MB/s when all four sticks are added to the MB (dual channel...1+3, 2+4).
Went back to BIOS and manually set the RAM settings to 5-4-4-9-30-3-4-3-4 to remove any AUTO RAM dectection mode. Re-tested both pairs. Still no change with the bandwidth numbers. BTW, all sticks passed Memtest86+.
Booted to Windows to check the performance of the RAMs using Super Pi. The 1st pair (higher bandwidth per Memtest) yields marginally faster time.
Can a single-sided RAM module be slower than a dual-sided stick? Is the BIOS having issue with proper detection of the single-sided modules? Has anyone seen a similar drop in memory bandwidth with 4 sticks vs 2 sticks? I'm aware that adding more RAMs will place a higher load on the memory controller. That's why most rigs will drop from 1T to 2T with four modules in place.
I also tested the single-sided sticks in dual channel without the double-sided modules on the board. I agree that the only explanation is the way the SB handle double vs single sided RAMs.
Does anyone encountered any drop in speed when moving from two sticks to four sticks?
Yes, it's normal for sigle-sided to perform worse than double-sided RAM because of the bank interleave, but the perfromance drop should be minimal.
Just shoved a cheapo E2140 in my IP35-PRO (its a stop-gap until the price-cuts), and decided to give it a little bit of a push, can probably go further than this, but seems to be stable so far :)
http://paradigital.net/XS/clock.jpg
EDIT:
Got it further now, 3200MHz
http://paradigital.net/XS/3200MHz.jpg
write speed should be better to a single sided row of memory chips...i would think with the single sided row stick you could lower your memory setting (make them tighter than the double sided) and in effect, get better or comparable read speed compared to the double sided.
It will be higher with higher density module. In my case, the dual-sided chip has twice the capacity of single-sided chip (same density). The single-sided chip can run at higher FSB speed. Do you have any recommendation for a free memory bench proggy to check these RAMs?
What are your pwm temps? in my board i think they are little high, always between 46-48º in idle, 56-57º in load :confused:.
for my board...i have 39-40C idle temps but i have 2 fans blowing on the heatsink over the PWM.
Thanks Ace;) , i try to put some air in that zone or remove the heatsink an put some fresh as5 to improve the contact.
Ace, im 400x8 now at stock board volts, whats your mobo volts to 450 fsb?
Make sure the stock heatsinc is actually touching the pwms. I took mine off and it was only touching about a third of them. I cut up a HDD cooler and put that on there and I see 36c idle and 38c to 40c load at 1.55vcore. With higher vcore it goes up from there where I ahve scene 53c before.
my vcore is between 1.53-1.55...
my room temp is 77F....AC
:eek: 36º idle and 40º load, this temps are much better. I don´t understand why abit don´t improve this little things. With a good contact heat-pipes this board are very cool compare with other p35´s.
Ace, 1,55v mch to 450?
Well this one I set up the phase on it before I even fired it up so I was hoping for the best. I kept the stock cooler intact so I could put it back on but I had a extreme spirit II that I put on the NB and modified my cooler for the pwm. Luckily it works like a charm. Great ocing mobo, cant wait to get a cpu that doesnt have a fsb wall at 540mhz to see how high I can get the fsb.
have you tried to uninstall and then re-install Uguru
What are ppl using to take a screen shot of uGuru ??
I have tried irfanview which works for everything except uGuru ????
No need for ALT. Just PrtScn, then open PAINT and EDIT/PASTE...FILE/SAVE. I also enjoy MWSnap...very light application. Works with all versions of windows.
@Ace-a-Rue & Turbodream : maybe you can try this.....:) :)
set lower tranparency setting on UGURU and restart.....I try this worked and still Ok...... I use Ati Card
http://img.techpowerup.org/070717/uguru.jpg
and when I Use Oc with Uguru and set PCI E 107 worked normaly.....no problem :) :)
http://img.techpowerup.org/070717/uguru 2.jpg
You must be pumping north of 1.5V with 3.62GHz. What's your peak Orthos/TAT/S&M temp? Do you have an L2 revision chip with 100C Tjunction?
My E4300 is semi-stable at 3.6GHz with 1.505Vcore and Big Typhoon/120 x 38mm Panaflo medium speed. Semi because I don't dare run TAT during the summer heat...80C within 30 seconds of test. Yes, the heat spreader has been lapped (top and bottom). I gained about 8C after re-mounting the heat spreader because of poor contact between it and the CPU core.
@furballi : I dont tell about OC performance but Screen UGURU on ATI Card :) :)
My Uguru works now, changed skin so now i can OC again :)
Serious question: I plan on getting q6600 and ds3r and I've heard good things about its OC abilities. Since I dont need SLI, should I stick with the p35-ds3r or get the IP35-pro?? :shrug: Its seems to me both boards are pulling off similar FSB.
does giga have something similar to Abit's Uguru?
What's so special about Uguru :confused:
you change FSB, PCIe bus speed and all voltages that count for overclocking right from windows...the only thing you can't do is change memory straps and multipliers...memory straps need a re-boot to accomplish a change in configuration!
this is what you can change from windows:
What is a good number for CPU GTLREF 1&3 and 0&2 ? Does it results in better fsb ?
i found out you can't change it to far from 67%, which is default on my board for my Q6600...once above 70%, my system becomes erratic...i do not know if different processors change those two numbers for default....i am hoping to get a readout from Fugger about what impact they really have on overclocking.
Not really - they've got easytune5 but imo their hardware monitoring/fan control is nowhere near as good & no POST reporter.
I use 70 on mine. Found that to be the most stable with running 1.24 cpu vtt. When I was running say 68 I would end up using a little higher cpu vtt but didnt do enough testing to really get a go of it.
I think these were posted here a while back but to put them up again.
Link1
Link2
Like I said I am not the one that brought this to anyones attention I am jsut throwing it back out there
i'm planning for my new system a qx6850 cpu and i am looking for a suitable board. i am not sure if IP-35 Pro would be a good solution for this or a Giga DQ6 would be cooler and better...anyone runnig a qx6850 on the abit?
greets from germany :)
i don't think that chip is out as of yet?
Everybody just started reviewing them and the other new 1333MHz FSB chips, but I haven't seen them out for sale either (nor have I seen the new price drops).
They did say 'July' for all that though, so I guess it will probably be sometime over the next two weeks.
that's my guess... :)
Well I have a first gen xifi fat
Would not work with my older nf4 board at all. Works fine so far with this setup, ip35 pro.
this not will work in vista though, on the same board.
I suspect drivers, or the fact i have a poo first gen, gonna RMA and see whats up. Thinking of just getting a bluegears
Noticed cpuz is showing massive droop. i realise its normal to show some, but this is bad. Alsu uguru shows quite a bit, under load it drops down to 1.38 1.39~~ish .
Should i be worried? I think i might be overloading one of my 12v rails.
i did test the 12v rail with DMM last month when this board was on the box, solid on the 12v
what could be causing this? its a ocz 700w gamextreme. should i be searching for a new psu. The system is 13 hours prime stable
http://www.imagehosting.com/out.php/i927551_droop.jpg
once you go above 1.350 (or so) set in bios you can't believe CPU-z. Look at what uguru says. That is more indicative of what is happening and actually that is even off as my board only shows about 0.05v difference between what is set in bios and actual (DMM).
I've noticed that CPU-Z & uGuru both get 'closer' in their readings if my CPU is under load (like 2 Prime95 instances).
i unplugged 4 fans. droop improved quite a bit. I will re route my fan connections.
the droop is about 1.40 ~ 1.41 now in uguru under orthos load. i read it was normal to have some droop with orthos anyway so i guess iam fine.
at idle though, yad think it would show 1.4! ohwell cant have everything
i know what you mean about droop.
1.55 = 1.48 for me.
I'm seeing about 0.04V droop with Abit IP35-E. E4300 @ 3.46GHz with TAT/S&M. 1.505 in BIOS. CPUz shows 1.448 Speedfan is around 1.46. C1E and EIST enabled in BIOS. PSU is Antec SP350.
so whats the overall concensus on this board? Thumbs up or down, especially on quad cores? Is the Asus better? I rather not spend the extra money. 8 phase vs 4 phase and junk like that.
Can anyone with a firewire drive confirm the poor throughput shown in this review?
bit-tech IP35 Pro review
What are the sata cable lengths included on these boards? Are they 19 inch or 24 inch?
About 5% of the time. S&M 1.9 is more reliable with WXP Pro SP1. Didn't work properly with Vista. If you need a reliable core measurement, then use Coretemp 0.95.