Abit IP35-E Review (500MHz FSB board)...$90 @ NewEgg + $6.61 ship

[Tweakin]

The sale @ the Egg runs out tonight so I'm going to pull the plug and pick one up to replace my aging MSI board running a tired X2 4200+.

I'm hoping to see a slight difference in performance...but I am sad about loosing my RAID.

 

[agent2099]

Originally posted by: nerp

Originally posted by: agent2099
Guys would this be a good budget overclockable setup?

ABIT IP35-E LGA 775 Intel P35 ATX Intel Motherboard - Retail
Intel Pentium E2160 Allendale 1.8GHz 1MB L2 Cache LGA 775 65W Dual-Core Processor - Retail
Crucial Ballistix 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Dual Channel Kit Desktop Memory - Retail

I'd say yes, but I'd hesitate on recommending the Crucial since so many people have had issues with them recently. You also want to get memory that is default 1.8v on this board. It has posed problematic for people trying to fire it up with 2.2v ram. I'm running 2x2GB G.Skill DDR2-800 1.8v and it runs like a TOP, I tell you.

The CPU choice is good. I got the E2180 for the 10x multi. 320 fsb x 10 = easy 3.2 ghz with only a small nudge in vcore.

If you plan on overclocking, I strongly recommend an aftermarket CPU cooler. The stock one that comes with the E2xxx series is pretty wimpy and doesn't have heatpipes or copper. I'm using a Tuniq Tower 120 on my board and there is plenty of clearance and room and load temps are at about 50C, which is quite good.

This is a fantastic board and one of the best I've ever owned. I think you'd be happy to join the community of IP35-E users.

Thanks. Would a 400W PSU be adequate for this setup with a 8800 GTS or 8800GT?


Also, would this be a worthwhile upgrade from a S939 system (Opteron @ 2ghz) ?

 

[Tweakin]

Also, would this be a worthwhile upgrade from a S939 system (Opteron @ 2ghz) ?

I think you will see some improvement, especially if your Opteron was a single cpu. I'm replacing my aging X2 4200+ system with this board and I expect to see a little gain, not much but a little.

 

[agent2099]

Originally posted by: Tweakin

Also, would this be a worthwhile upgrade from a S939 system (Opteron @ 2ghz) ?

I think you will see some improvement, especially if your Opteron was a single cpu. I'm replacing my aging X2 4200+ system with this board and I expect to see a little gain, not much but a little.

My Opty is a 165 Dual core, About an identical processor to your X2 4200+.

Hmm I'm starting to have 2nd thoughts based on what you said. I don't even game much, mostly thinking of upgrading because I multitask quite a bit.

 

[Tweakin]

Originally posted by: agent2099

Originally posted by: Tweakin

Also, would this be a worthwhile upgrade from a S939 system (Opteron @ 2ghz) ?

I think you will see some improvement, especially if your Opteron was a single cpu. I'm replacing my aging X2 4200+ system with this board and I expect to see a little gain, not much but a little.

My Opty is a 165 Dual core, About an identical processor to your X2 4200+.

Hmm I'm starting to have 2nd thoughts based on what you said. I don't even game much, mostly thinking of upgrading because I multitask quite a bit.

Remember, a 2.5Ghz Intel (C2D) is going to be faster then a 2.5Ghz AMD Opty/X2. By the time you ramp your cpu up to 3Ghz, or buy one that will do 3Ghz, the difference should be noticable, how much is up to each users opinion...I'm being driven as I have a bad IMC on my X2 and I'm not going to put any money in a dead (939) avenue. I'm also looking forward to lower temps in my case and more silent operation.

Maybe you should look into a quad core then...

 

[agent2099]

You are right I thought about Quad Core. But the fact that I'll also be doubling my RAM in addition to getting a C2D I figured I'll go ahead and see what this improvement gives me first.

I went ahead and bit just now. Can't fight the upgrade bug .

I figure if I need more speed down the line then Quad core is an option and since I'm in the C2 arena now I can just drop another chip in, instead of upgrading multiple components. And by then Quad core prices may have dropped even further.

 

[leoQC1]

A big thank you for this thread!! Over the past week while waiting for my E6600 I got for a steal on FleaBay, I was obsessing over my shiny new IP35-E and found this thread while doing some research. Thanks to the wealth of experience shared here, I not only steered clear of OCZ ram (Serpent, but no N5"s sorry), but had a trouble free installation, knowing what to expect. Previously I was running a "family rig" of E6300 on Gigabyte 945P-DS3 with Hynix Ram, and have been blown away by the speed and simplicity of this board. I used to swear by my Abit NF7 in the Athlon is King days but went to Gigabyte after Abit got hammered in forums left, right and centre. Man, it is nice to be back with Abit. I found the Gigabyte to be a truly temperemental piece of crap, which was a surprise after so many swear by them. It used to run nice for days then have BSODs, random reboots etc, with no discernable cause then go back to working fine again.
I have had my E6600 cruising nicely at 3.0gHz while I run it in for a couple of days (so easy to overclock) and the IP35-E is stable as. I use Prime and Memtest, but after this thread will try Orthos too. Once I am happy all is well I will try going North and see what I can extract out of this baby. After what I have read here I am hoping for some good results.

One question

The NB heatsink appears to be rather hot (though I can touch it ok) but AbitEQ reports it as 49 (idle)(CPU=31). I checked the sink face before installation and all looked ok, even though the sink itself moves quite easily, when prodded. Am I worried about nothing? or do you suggest changing to an active or tower sink if overclocking? the SB and Mosfet sinks are only warm compared to it.

Thanks for the thread once again.

 

[Heidfirst]

Originally posted by: leoQC1
The NB heatsink appears to be rather hot (though I can touch it ok) but AbitEQ reports it as 49 (idle)(CPU=31).

abitEQ doesn't cover the NB, there is no sensor.

 

[Tweakin]

Originally posted by: leoQC1
The NB heatsink appears to be rather hot (though I can touch it ok) but AbitEQ reports it as 49 (idle)(CPU=31). I checked the sink face before installation and all looked ok, even though the sink itself moves quite easily, when prodded. Am I worried about nothing? or do you suggest changing to an active or tower sink if overclocking? the SB and Mosfet sinks are only warm compared to it.

Thanks for the thread once again.

I would recommend that you try removing the sinks and applying some better TIM and see what happens. I always do this first thing as I have never found TIM properly applied from the factory.

 

[kalster]

is anyone using an esata bracket with this board, i bought the antec mx-1 that came with a esata bracket,i have it connected to one of the sata ports on the motherboard, whenever my antec-mx1 is powered up , the bootup takes forever, it takes a long time to recognize the drive connected to the esata port (i tried changing sata ports on the motherboard too but it did not help), also is there a way to setup a sata port as hot pluggable

 

[QuixoticOne]

Originally posted by: Tweakin

Originally posted by: leoQC1
The NB heatsink appears to be rather hot (though I can touch it ok) but AbitEQ reports it as 49 (idle)(CPU=31). I checked the sink face before installation and all looked ok, even though the sink itself moves quite easily, when prodded. Am I worried about nothing? or do you suggest changing to an active or tower sink if overclocking? the SB and Mosfet sinks are only warm compared to it.

Thanks for the thread once again.

I would recommend that you try removing the sinks and applying some better TIM and see what happens. I always do this first thing as I have never found TIM properly applied from the factory.

QFT. If you really want to improve the thermal situation (use FULL ESD precautions and be also mechanically very careful) take off the NB sink, scrape off the thermal pad / wax / whatever it has on the chip and heatsink, use pure IPA or something similarly appropriate like Arctic's heatsink cleaner to clean the heatsink and chip, then apply some Arctic Ceramique or Arctic Alumina or maybe Arctic MX-1 to the chip and reinstall the HS.

Consider shimming or whatever to get the heatsink to clamp down a little more firmly if that's easily possible, otherwise be prepared to apply a decently generous amount of paste so it'll make contact even after it shifts around a bit.

Or you could just put on a NB heatsink+fan that's made for the spot.

As for the MOSFETS, I'd be a little more cautious about reinstalling the heatsink / TIM on those until you've figured out if you can even get the heatsink to stay put and stay on straight / moderately tightly and make contact with all the chips underneath etc. Sometimes they just don't fit well and you really kind of need thick TIM or something.

I'd get a 60mm or 90mm fan and have it blow across the NB and another to blow across the mosfet / motherboard power converter area.

 

[Heidfirst]

Originally posted by: kalster
also is there a way to setup a sata port as hot pluggable

officially ICH9 only supports AHCI under Vista (although I believe that it can be hacked under XP) but does the drive support hot-plugging as well?

 

[imported_Innovato]

Thanks for the great review - I ended up picking one of these up. Question: did you use the supplied SATA cables with your hdd? I bought a Corsair VX550W and a Western Digital SE16 640 GB and cannot plug both the SATA and power plugs in at the same time - not sure if the Abit SATA cables are a tad wider than standard. Sad story short, the hdd was the last component of my build and trying to plug the two cables in resulted in the plastic of the hdd snapping off...in all the years of building computers, I can't say I've ever felt a hdd feel so cheaply made...

Any thoughts/advice?

Thanks!

 

[QuixoticOne]

Originally posted by: Innovato
Thanks for the great review - I ended up picking one of these up. Question: did you use the supplied SATA cables with your hdd? I bought a Corsair VX550W and a Western Digital SE16 640 GB and cannot plug both the SATA and power plugs in at the same time - not sure if the Abit SATA cables are a tad wider than standard. Sad story short, the hdd was the last component of my build and trying to plug the two cables in resulted in the plastic of the hdd snapping off...in all the years of building computers, I can't say I've ever felt a hdd feel so cheaply made...

Any thoughts/advice?

Thanks!

My condolences on your HDD. I just hope it broke BEFORE you had any useful data on it.

Whoever designed the SATA interface connectors / cables should be taken out and hung by a tree with SATA cables or something.

I bought about 5 different lengths, thicknesses, connector styles of SATA cables and whenever I need one I always dig through the box to find one that actually almost fits. It is insane like Goldilocks and the three bears..

"No, this one is too long..."

"No, this one is too short..."

"No, this one is too thick..."

"No, this one is too flimsy..."

"Ah! Wonderful, this one ALMOST works right without falling off or being about to break the connector off the hard disc or having to be origami looped around in kama suttra ways!"

I think some of the cables even have a little locking pin that is different than usual so you can only use those cables with certain motherboards or drives or something to help them stay on at the right angle without torquing off or whatever... I know I read about that incompatible design "feature" in something I bought or was going to buy recently... maybe it was the WD hard drives... maybe it's the ABIT IP35-E.. I forget and haven't assembled them to find out which it is yet.

Anyway other than being on the lookout for non-standard connector "features" like locking pins, you really just have to use whatever SATA cable actually FITS without too much stress...

Big tip -- a good tip is to get some zip-ties or wire-ties or velcro-ties or whatever and use them to tie down the sata cable at one or two points where it HAS to be at the right angle, the right stress level, with the right bend radii, and with the right routing in the PC -- otherwise it'll be a disaster and either work loose or be too tight or get in the way or something.

Also watch out for data errors, SATA cables don't really like being too close to anything else since they induce and are susceptible to electrical noise / disruption by radiating to other metallic objects / cables / case parts if they get too close for too long of a distance at too parallel of an angle.

 

[Spammeh]

Have I got a bad overclocking q6600 guys? :[

I can't seem to do 333x9 with stock volts or even up to 1.36v. It fails Prime small fft in a few mins. I have a vid of 1.3250v. It's prime small fft and blend stable at stock.
I've made sure my ddr is running at rated speeds, volts and timing.

Temps are 45c-42c under load from core0 to core3 (stock speeds)

any ideas? :[

 

[Tweakin]

Originally posted by: Spammeh
Have I got a bad overclocking q6600 guys? :[

I can't seem to do 333x9 with stock volts or even up to 1.36v. It fails Prime small fft in a few mins. I have a vid of 1.3250v. It's prime small fft and blend stable at stock.
I've made sure my ddr is running at rated speeds, volts and timing.

Temps are 45c-42c under load from core0 to core3 (stock speeds)

any ideas? :[

Maybe...did you make sure to set your memory to 1:1 and not Auto? Auto will cause the memory to overclock and this may be holding you back.

I would also try to back the fsb down a bit and keep trying, maybe even loosen up your memory timings a bit.

 

[mindless1]

Originally posted by: Spammeh
Have I got a bad overclocking q6600 guys? :[

I can't seem to do 333x9 with stock volts or even up to 1.36v. It fails Prime small fft in a few mins. I have a vid of 1.3250v. It's prime small fft and blend stable at stock.
I've made sure my ddr is running at rated speeds, volts and timing.

Temps are 45c-42c under load from core0 to core3 (stock speeds)

any ideas? :[

You didn't tell us about any other settings, does it mean you didn't change others?

Don't be afraid of minor voltage increases, I suggest you increase all voltages (except southbridge which shouldn't need any) one notch and see if that helps. When you appear to have regained stability, lower the voltages again as much as prudent.

Offhand I would try 2.0V for memory and 1.33V MCH. While 1.36V vCore is higher than most people need, go ahead and try around 1.40V, this change isolated from the others so if it doesn't help you can lower it again.

 

[Spammeh]

I'm using the ratio 1:1.20? can't remember off my head but it's keeping the memory at 798mhz. Should I still loosen my memory timings if they're at spd timing?

I kept VTT, ICH, MCH at stock voltages since i keep hearing of people not needing to adjust them at low fsbs like 333.

At 333x9 with 1.32vcore (stock) - small fft will fail on one core with a rounding error
- Increasing vcore up one notch and two (1.36) just makes it reboot instead of displaying the error, what does this mean? heat problem at nb perhaps? I can see the temps are good otherwise.

http://img138.imageshack.us/img138/5290/tempsdb7.gif

 

[MustangSVT]

God damn it.... it was working perfectly fine for few days.

IP35-e
e2160 @ 333 fsb = 3.0ghz
2gig of that HP ram
antec 380w earthwatt psu
and 8800gs

voltage was stock on all. and latest 1.6 bios.

it was working fine without any dual boot issues. yesterday I thought I would try to change the timing on ram so I went into settings and tried 4.4.4.4.15 which shouldn't be a problem at stock 333fsb.
but that was the start of the problem..

after that It would not bootup. it would either reset it self after booting up or would do a long beep pause and long beep...
it would not display anything on screen.

It would continue to do so even after cmos reset and draining power from mobo.
after several tries, it would sometimes boot up ok but when I try to oc it would have this problem again.

What's odd is that it has been working/running fine for about 4 days without any issue. no new components were added nor any changes in system. it was o/ced to 345fsb and ran without problem before.

I will try to go through this thread again but, what are some things I can try? its really odd that its acting up when nothing has been changed and bios is completey reset

arggggggggggggggggggggg

 

[mindless1]

Originally posted by: MustangSVT
God damn it.... it was working perfectly fine for few days.

IP35-e
e2160 @ 333 fsb = 3.0ghz
2gig of that HP ram
antec 380w earthwatt psu
and 8800gs

voltage was stock on all. and latest 1.6 bios.

it was working fine without any dual boot issues. yesterday I thought I would try to change the timing on ram so I went into settings and tried 4.4.4.4.15 which shouldn't be a problem at stock 333fsb.
but that was the start of the problem..

after that It would not bootup. it would either reset it self after booting up or would do a long beep pause and long beep...
it would not display anything on screen.

It would continue to do so even after cmos reset and draining power from mobo.
after several tries, it would sometimes boot up ok but when I try to oc it would have this problem again.

What's odd is that it has been working/running fine for about 4 days without any issue. no new components were added nor any changes in system. it was o/ced to 345fsb and ran without problem before.

I will try to go through this thread again but, what are some things I can try? its really odd that its acting up when nothing has been changed and bios is completey reset

arggggggggggggggggggggg

I have an E2180 CPU, it does not like bios 16 at all so I stick with bios 15. When I needed to clear CMOS I unplugged AC and used the jumper but for some reason it didn't clear it, I then unplugged AC, pushed the power switch on the case front, pulled the battery for 10 minutes and used the clear CMOS jumper. Success was indicated by the system clock being reset which it wasn't previously.

Your mileage may vary, but I would try these things. I would also try raising vcore slightly at 3GHz, just because that works when system is brand new doesn't necessarily mean it left enough margin that it still works after the system ages a little. Remember we're talking about mere tenth(s) of a volt difference.

Also, If you're talking about the DDR2-667 HP memory, I'm running 4 sticks on this board (maybe causing more conservative settings to be needed vs 2 sticks) and would not think you're assured to run ok at 4,4,4,15. Mine is at 5,5,5,15:2T with 1:1.25 ratio at 333FSB and 2.0V - I was too lazy/lack-of-time to bother testing at lower voltage though others have had success lower. With 4 sticks mine will run ok up to roughly DDR2-850 speed at aforementioned timings.

Try MCH at 1.33V, there's no reason to refrain from moderate voltage increases to see if it resolves some problem - if it doesn't just return them to stock again.

Every part has a minor variation, what works for myself or others may not be exactly what you need. If memory problems keep you from getting a stable post to change settings again, try leaving only one memory module installed temporarily.

 

[mindless1]

Originally posted by: Spammeh
I'm using the ratio 1:1.20? can't remember off my head but it's keeping the memory at 798mhz. Should I still loosen my memory timings if they're at spd timing?

I found 1:1.2 quirky, try 1:1.25

I kept VTT, ICH, MCH at stock voltages since i keep hearing of people not needing to adjust them at low fsbs like 333.

If someone else doesn't need to change them, it's not necessarily proof you don't. Raising MCH a notch or two is quick, easy, painless. Just try it. If it doesn't help return to stock voltage. Leave it a notch or two higher till you achieve stability THEN lower it if possible.

At 333x9 with 1.32vcore (stock) - small fft will fail on one core with a rounding error

After how long? Immediately would tend to mean voltage too low, after more than a few seconds it could be voltage in conjunction with higher temps (not instabily high necessarily but higher temp cases chips to need more voltage too).

FWIW, mine (E2180) had stock vcore of 1.3125, but the resulting actual voltage as displayed in CPU-Z or speedfan was 1.284 or so. I found I could run up to about 2900MHz at stock voltage, but closer to 3000MHz I'd have a reset or cold boot problem about every 4th time if voltage wasn't raised to 1.3325 in bios. At 3070MHz or so it needed 1.3525V. YMMV, just remember some people get better than average chips, some worse than average with respect to o'c potential, max FSB, and vcore at any speed.

- Increasing vcore up one notch and two (1.36) just makes it reboot instead of displaying the error, what does this mean? heat problem at nb perhaps? I can see the temps are good otherwise.

Why would changing vcore and seeing a difference indicate heat at another chip? Since it makes a difference it would seem to indicate you need to go further raising vcore, but if you haven't observed enough instances of the problem it could be random which happens. Remember these chips have thermal shutdown, it will not melt if you went to 1.40V, or even slightly higher. Just remember to monitor temps when doing so, if an Orthos test pushes it past 70C you might as well quit the test assuming you need to lower voltage next, and clockspeed too if nothing else works.

Remember that to isolate things you don't have to guess, you can lower multiplier on CPU and keep same FSB and memory ratio to see if that makes a difference. Also temporarily disable C1E and EIST in the bios if you hadn't yet, leave them like this until stability is achieved and THEN re-enable and retest, same as with trying lower voltages you'd want to do the most conservative things and leave lots of margin then selectively try tweaking each setting in turn. At 3GHz you have lots of thermal margin left to try more aggressive voltage, till you know if it makes a difference.

 

[manimal]

I switched out the processor in the IP35E rig in my sig to a E8400. I am getting more Vdroop at load with the e8400 than I got with my Q6600 oddly. PWM temps are down dramatically so I can boot and bench at 450-470 fsb. The problem is is with the Vdroop I have to set Vcore at 1.505 to get 1.4 at prime load which makes me stable at 4.0. I am using Micron D9 rams that do 1200 on this rig and others. I have a HR05 on the NB which remains cool to the touch even after a hour at prime small ffts load and I have a VCRe on the SB. PWM temps are in the 40s at prime load small ffts as well so its not in anyway temp related. My e8400 has 1.0375 so from what ive read it should be a decent chip. I may switch out the board with a DS3L I got at frys in this bundle but I love my IP35E and ive put in a ton of work-lapped the pwm sink, nb amd sb coolers-etc...


Anyone else getting that big of a droop?
voltages at 4.0 are all stock execpt ram 2.0 Vcore 1.505 bios 1.4 load actual

 

[Spammeh]

Thanks for the detailed response mindless1!

I've set the ratio 1:1.25 instead, my ram is running at 832mhz

bumped vtt,ich, mch up a notch each.

vcore still at stock - lol prime crashed but without rebooting machine?

Usually it reboots with 5-6mins of prime small fft.

http://img135.imageshack.us/im...9923/primecrashrs5.jpg

 

[MustangSVT]

Originally posted by: mindless1

Originally posted by: MustangSVT
God damn it.... it was working perfectly fine for few days.

IP35-e
e2160 @ 333 fsb = 3.0ghz
2gig of that HP ram
antec 380w earthwatt psu
and 8800gs

voltage was stock on all. and latest 1.6 bios.

it was working fine without any dual boot issues. yesterday I thought I would try to change the timing on ram so I went into settings and tried 4.4.4.4.15 which shouldn't be a problem at stock 333fsb.
but that was the start of the problem..

after that It would not bootup. it would either reset it self after booting up or would do a long beep pause and long beep...
it would not display anything on screen.

It would continue to do so even after cmos reset and draining power from mobo.
after several tries, it would sometimes boot up ok but when I try to oc it would have this problem again.

What's odd is that it has been working/running fine for about 4 days without any issue. no new components were added nor any changes in system. it was o/ced to 345fsb and ran without problem before.

I will try to go through this thread again but, what are some things I can try? its really odd that its acting up when nothing has been changed and bios is completey reset

arggggggggggggggggggggg

I have an E2180 CPU, it does not like bios 16 at all so I stick with bios 15. When I needed to clear CMOS I unplugged AC and used the jumper but for some reason it didn't clear it, I then unplugged AC, pushed the power switch on the case front, pulled the battery for 10 minutes and used the clear CMOS jumper. Success was indicated by the system clock being reset which it wasn't previously.

Your mileage may vary, but I would try these things. I would also try raising vcore slightly at 3GHz, just because that works when system is brand new doesn't necessarily mean it left enough margin that it still works after the system ages a little. Remember we're talking about mere tenth(s) of a volt difference.

Also, If you're talking about the DDR2-667 HP memory, I'm running 4 sticks on this board (maybe causing more conservative settings to be needed vs 2 sticks) and would not think you're assured to run ok at 4,4,4,15. Mine is at 5,5,5,15:2T with 1:1.25 ratio at 333FSB and 2.0V - I was too lazy/lack-of-time to bother testing at lower voltage though others have had success lower. With 4 sticks mine will run ok up to roughly DDR2-850 speed at aforementioned timings.

Try MCH at 1.33V, there's no reason to refrain from moderate voltage increases to see if it resolves some problem - if it doesn't just return them to stock again.

Every part has a minor variation, what works for myself or others may not be exactly what you need. If memory problems keep you from getting a stable post to change settings again, try leaving only one memory module installed temporarily.

weird. I am still baffled by what is causing this problem.

I cant seem to focus it down to one issue.

I've tried 1 stick of ram on all four slots which doesnt seem to make any difference.

tried to give more voltage to all things and still does it randomly.

If I set fsb to anything above 233 it seems to give me those issues.


And I've tried to run uGuru thing that game with abit hoping to o/c within windows and it just comes up blank.. any ideas?


i'm still going thru the thread but at least I have it up @ 233fsb

could it be the temperature?
is 16 bios bad?

mmmmmmm

ahhhh is uguru ONLY for ip35-pro? damn abit.. threw me off by having it on the install cd for -e.

what other prog can i use to o/c within windows?

 

[lopri]

a