Overclock.net banner
1 - 15 of 15 Posts

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #1 ·
Last evening, i took the plunge and decided to swap my Gigabyte 970 UD3P with my Biostar TA 970 Plus (6+1 phase). Oh, pretty!



Installed Windows, all fine. Well, almost. The BIOS although pretty, had some strange names. Like:

- VGA palette snoop (never seen this before)
- PCI-E Maxy Payload
- ASPM
- Cool N Quiet apparently is called "PowerNow".
- Core levelling (explanation saying "core downclocking") i still don't know what it is.
- CPM Mode
- CSM 16 Parameters
- INT9 Trap
- Gate 20 Active
- Turbo core is apparently called "CBP".

Anyway, all seemed fine, until i installed monitoring software. Now in BIOS there was only 2 values, CPU and system, both very low, like 32C. The same 2 values are the only that show in Biostar's monitoring software, which is as spartan as it gets. It shows just these 2 temps and nothing more.

To cut to the chase, you have a 3rd temp in all other monitoring programs, which as you boot to Windows is already at something like 64C and as you stress test, it hits 90C and it throttles.





With Prime95 it throttles even at stock 3.5Ghz, as soon as that temp hits around 90C. So, the suspicion is, that it's the VRM. The VRM heatsink, measured with IR thermometer, shows temps between 47C and 58C. So i don't know what's going on, unless the thermal pad down there is completely crap and isn't transferring any heat.

Mind you, these temps, with a top-down cooler with 120mm fan at 1900rpm. Speaking of which, the fan's speed doesn't seem to be able to impact this temperature, as you see 1200rpm on one and 1900rpm on the other.

The CPU auto fan took me a while to figure out, which is why this discrepancy. It's actually INGENIOUS. You press a button and the motherboard "calibrates" your fan and finds min rpm, max rpm and how it revs up with voltage. At the end, it proposes you values to input manually and you get 2 autoprofiles, "quiet" and "aggressive". And they both work. It's the best method i 've seen between Gigabyte and Asrock.

To add insult to injury, if you attempt to overclock to 4Ghz, which can be done only by an option about "custom P-States", you lose Cool N Quiet, despite Power Now being enabled. I mean, PULEEASE! This is so 2007!!!

The audio is also very clear on the motherboard and very loud.

Maybe it's PEBKAC, but i really can't figure out what else i might have done wrong. Maybe it's a defective part, i don't know, but now i 've derated it to 0.5 out of 5 stars, in my previous review:

http://www.overclock.net/products/biostar-ta970-plus/reviews/7112

It's a SHAME really, because everything works very well (including the USB 3 ports). BUT, for a 6+1 phase motherboard, to throttle at stock clocks and actually it undervolts too, due to vdroop to 1.27v and to lose Cool N Quiet, is inexcusable. Could be that the VRM heatsink thermal pad is completely junk and if replaced, everything goes to normal. I don't know. But i don't have the time or patience to try to fix it.

The good thing is, i didn't pay much for it. I will keep it as spare for my brother or some friend probably. And now i must decide whether to put back the Gigabyte UD3P with the terrible BIOS or put back one of my spare Asrock 970 Extreme3, which at least can withstand 4Ghz without throttling and have easygoing BIOS.

Anyway, this explains why Biostar doesn't sell many motherboards. This could have been a winner, but they ruined it.

EDIT: I pressed down hard the VRM heatsink and i found out that it can be easily displaced actually. This also lowered said temperature to 44C, but under stress test it went up to 90C the same. So, the cuprit may be the bad heatsink design/ bad thermal pad / not enough pressure of the heatsink / poor VRM design.

Any of that or all of them combined. Fact is, i am exhausted and don't have energies for trying to fix it. I have Arctic thermal pad i could use, but this would require me to take it off the case and reinstall it again and if Arctic doesn't work either, take it out again. Which i just can't stand.

I think i will put an Asrock 970 Extreme3 back. I need a break from Gigabyte too.

P.S: As a bonus, the Biostar manual doesn't explain ANYTHING about the BIOS. I mean, ANYTHING. They think it's easy to guess.
biggrin.gif
 

·
Registered
Joined
·
5,625 Posts
i had a TZ77XE4 and it was rock solid.

would buy again.

most of those unknown temp's are a dead sensor on the board...

is your north bridge heatsink hot?
 

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #3 ·
Quote:
Originally Posted by Master__Shake View Post

i had a TZ77XE4 and it was rock solid.

would buy again.
This one is rock solid too. It just overheats. It's easy to make a "rock solid" motherboard in Intel. Cause the demand from the VRM is low. But where the game gets tough, it's in AMD.

Intel is like running an FX4300 on Asrock 970 Pro2. It will run it without any problem. Try an 8350 and things change radically.
 

·
Registered
Joined
·
600 Posts
Did you update the bios ?

How about replacing the thermal pads under vrm heatsinks with good ol' thermal paste ?

Also, just fyi, number of phases doest not necessarily means good power delivery. You can have many phases and they're crap.
 

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #6 ·
Quote:
Originally Posted by hojnikb View Post

Did you update the bios ?

How about replacing the thermal pads under vrm heatsinks with good ol' thermal paste ?

Also, just fyi, number of phases doest not necessarily means good power delivery. You can have many phases and they're crap.
1. There's only 1 BIOS released (the initial one). The motherboard is pretty new, when it was released, all AMD CPUs were already supported. Before changing anytning, i loaded optimized defaults, because there were settings i couldn't figure out. From there, i modified the stuff i knew and needed.
2. Thermal paste will only make the contact with the heatsink even worse, because the VRM heatsink is held in place with springs. Removing the pad for paste, would only make the heatsink more loose. I do have 1mm thick Arctic thermal pad, but i don't have the energy to get the motherboard out, change pad, put it back in and if it still overheats, put it out again. I have plenty of spare motherboards, i will just use another.
3. I know, but this is certainly a first. It can't even tolerate an FX at 3.5Ghz. I mean, mercy, it's supposed to be in the official support list, it boasts 140W CPU and can't cope with UNDERVOLTED FX at 3.5Ghz.

I will keep it for my brother, who runs low TDP Athlons in his office and if times comes, before i install it, then i will change pads.
Quote:
Originally Posted by ShrimpBrine
Ive bricked a couple few Biostop boards in the past. The just cant hang. Never used for a daily rig though. But if thats all there is to use, better than nothing or 4+1 vrm mitx.
I believe you...
 

·
Head Smeghead of OCN
Joined
·
3,159 Posts
I have a TA990FXE that worked well until it ran into a problem with it being used to test a potentially bad stick of RAM.
The board right after turning it on shut off and that was it - The board now acts as if there is no power to it at all as if the PSU isn't plugged into the wall. This same stick of RAM was also tested in one of my Asus boards and all it did was cause a RAM error, the board is fine and working to this day.

I have to say if a simple bad stick of RAM can do this to a board while another brand can handle it, makes me wonder about the quality of the boards in general but at least this one did work just fine as a DD setup until I ran into the bad stick and tested the piece.
If I knew what to look for on the board I could probrably fix it and put it back in service, haven't seen many if any of these around lately.
 

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #8 ·
Quote:
Originally Posted by Kryton View Post

I have a TA990FXE that worked well until it ran into a problem with it being used to test a potentially bad stick of RAM.
The board right after turning it on shut off and that was it - The board now acts as if there is no power to it at all as if the PSU isn't plugged into the wall. This same stick of RAM was also tested in one of my Asus boards and all it did was cause a RAM error, the board is fine and working to this day.

I have to say if a simple bad stick of RAM can do this to a board while another brand can handle it, makes me wonder about the quality of the boards in general but at least this one did work just fine as a DD setup until I ran into the bad stick and tested the piece.
If I knew what to look for on the board I could probrably fix it and put it back in service, haven't seen many if any of these around lately.
Yeah, it does seem that the BIOS has some "evil quirks". Like, the first time that i tried to install Windows, i loaded optimized defaults and didn't notice that it sets SATA to IDE. So installed Windows in IDE mode. When i realized that, i changed it to AHCI. Now normally, Windows will allow you to change from IDE to AHCI. But the motherboard went bananas and it would blue screen before entering Windows and going to an endless loop. THe blue screen was appearing for a fraction of a second and it was just rebooting. I dind't have the time to read what the screen was saying. So i had to install a 2nd time with SATA as AHCI and then it all went fine... Until i installed the monitoring software...

I went to sleep at 03:30, i was at work at 07:00, i am destroyed. I want to to pull it out now and can't find the strength.
 

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #9 ·
OK, there is definitely the thermal pad at least partially at fault here. I removed the motherboard and like i said, the VRM heatsink can be easily displaced, so i moved it to the back and looked at the thermal pad. It's an ultrathin blue pad, with parts of it stuck on the mosfets and parts stuck on the heatsink. So it's in bad shape.

At any rate, i put a spare Asrock 970 Extreme3 in and...despite the fact that my 1st such board (now on my secondary rig) had run [email protected] for months without issues, this one can't. It throttles at about half way in IBT AVX, reaching 68C CPU temp. I don't know what to say... Maybe i should start replacing every thermal pad i find...

At least it seems to have no problem for the stock 3.5Ghz... CPU temp is acceptable...



I am exhausted, i have yet to reinstall Windows, cause now i have Biostar drivers running and i can't have this, don't have the strength to put back the GIgabyte UD3P, so i ll stay with the Asrock at 3.5Ghz.

The irony is that this Asrock has never been used. Its sister that would happily run 4Ghz at 62C temp, had taken a beating for some years...
 

·
woot
Joined
·
2,623 Posts
vga palette snoop...thats a blast from the past...it was necessary when running a pci decoder or tv card that worked in conjunction with the graphics card. last time i remember using it was with a hauppauge tv tuner and an agp vid card
 

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #11 ·
Quote:
Originally Posted by sli_shroom View Post

vga palette snoop...thats a blast from the past...it was necessary when running a pci decoder or tv card that worked in conjunction with the graphics card. last time i remember using it was with a hauppauge tv tuner and an agp vid card
There were plenty of settings that i 've never seen before neither in Asrock nor in Gigabyte BIOS. What you say goes well with the inability to overclock and keep **** N Quiet active. This was a characteristic of AM3 motherboards. I have yet to see an an AM3+ that can't do that.

Anyway, here's the difference with the Asrock 970 Extreme3 (4+1 phase, originally supporting FX8350, later removed from CPU compatibility list).

I added a 2nd case fan, removed the Scythe Rasetsu for the Scythe Katana3 to make room for the case fan, added a 6mm fan on northbridge (couldn't fix it on VRM area). And 4Ghz passed at Very High.





But in a couple of months i will buy some other motherboard. I don't trust this for summer temps. Now it's 24C ambient, which is quite cool. I don't want to use the UD3P.
 

·
curmudgeon
Joined
·
5,933 Posts
That's a shame about the Biostar. I was looking into their top model when they quit selling it. It had a decent reputation at that time. (2013)
 

·
Premium Member
Joined
·
4,607 Posts
Discussion Starter · #13 ·
Quote:
Originally Posted by miklkit View Post

That's a shame about the Biostar. I was looking into their top model when they quit selling it. It had a decent reputation at that time. (2013)
It's a TERRIBLE shame, because other than the VRM skyrocketing (even at idle, i was getting 60C), the board worked without a hitch. The USB3 ports all well, the audio VERY loud (it has headphone amplifier on). But, the VRM and the loss of Cool N Quiet killed the whole experience. Before, i was CERTAIN that at LEAST it would beat the Asrock 970 Extreme, cause you have 2 more CPU phases, how bad can 6 phases be! Turns out, they can be pretty bad...

Part of the problem, is certainly the thermal pad. But i don't have the patience to fix and try it again. At some point, i will order many Arctic thermal pads and i will replace ALL pads on ALL my motherboards, PREEMPTIVELY.
 

·
Head Smeghead of OCN
Joined
·
3,159 Posts
Quote:
Originally Posted by Undervolter View Post

Yeah, it does seem that the BIOS has some "evil quirks". Like, the first time that i tried to install Windows, i loaded optimized defaults and didn't notice that it sets SATA to IDE. So installed Windows in IDE mode. When i realized that, i changed it to AHCI. Now normally, Windows will allow you to change from IDE to AHCI. But the motherboard went bananas and it would blue screen before entering Windows and going to an endless loop. THe blue screen was appearing for a fraction of a second and it was just rebooting. I dind't have the time to read what the screen was saying. So i had to install a 2nd time with SATA as AHCI and then it all went fine... Until i installed the monitoring software...
To clarify the problem is hardware in nature, not software or BIOS related.

When I say you go to turn the PSU on to the board/system I literally mean that from turning on the PSU itself with it's switch - NO standby LED's or anything light up to even show a sign of life in it. Same thing as if the system were unplugged from the wall outlet and that's exactly what it does - Nothing.

This really was a good board right up until that happened to it.
 

·
Registered
Joined
·
2,725 Posts
Quote:
Originally Posted by undervolter
i will order many Arctic thermal pads and i will replace ALL pads on ALL my motherboards, PREEMPTIVELY.
I don't know of any pads with a better reputation than Fujipoly.
 
1 - 15 of 15 Posts
Top