Overclock.net › Forums › AMD › AMD Motherboards › MSI A88X-G45 GAMING - FM2+ goodness
New Posts  All Forums:Forum Nav:

MSI A88X-G45 GAMING - FM2+ goodness - Page 6

post #51 of 99
Quote:
Originally Posted by luminousone11 View Post

Their is some funny business in the A88X g45's vrm mosfets. I have noticed the same voltage fluctuation problem albeit, not to the same extreme, I am starting to wish I picked up the gigabyte up4 board.

Pictures of the mosfets on the msi board show 22 mosfet chips, But I have yet to find a picture that I can make out the numbers on the chips with, At first I though the board was using a doubler, But the more I read from real world usage of the thing the more I dough that.

I believe it used to be, that it's 4 chips: 1 FET Driver each for High and Low sides, in addition to the High and Low side MOSFETs, meaning 4 chips per phase. As such you count the chokes and that's how to tell how many phases it actually has. Though I seem to have found that with ASRock they tend to use twice as many smaller chokes, but you can still tell by how many caps it has on the output side (right before the CPU).

Thing is, phases have gotten better and more efficient so we don't NEED 16 on AMD or 24-32 on Intel anymore. Still, I would've liked to see 4+2+2 at least (CPU+GPU+IMC)
Edited by Formula350 - 3/21/14 at 9:47pm
post #52 of 99
Quote:
Originally Posted by luminousone11 View Post

Pictures of the mosfets on the msi board show 22 mosfet chips, But I have yet to find a picture that I can make out the numbers on the chips with.

Only thing I have found is the reference to the Intersil ISL6377 IC which is "specifically designed" for AMD APUs. I tried looking into the spec sheet on it, but I got a little lost. I posted another video with the DVOM readings here: https://www.youtube.com/watch?v=LgzmjB9KBjo - I know it's not the best meter but it still a big fluctuation.
Quote:
Originally Posted by Formula350 View Post

Since your chip is so damn hot it could easily be heating up the VRM sinks a bit. Also with as much copper as the boards have they will spread the heat around more as well, which will make nearby components heat up a bit further.

That's an excellent point right there, makes perfect sense! The sinks are pretty close proximity too.
Quote:
Originally Posted by Formula350 View Post


As for the sound, I have not tested it, sorry. I use my 46" LED TV as my monitor, and so I use the HDMI Audio and my TV"s Line-out into my speaker system for sound. Even then, I have a high end sound card which pummels the onboard's abilities lol I've never found Realtek's offerings to be very good at the low frequencies though, but then again last time I used one was back before they were actually giving them suitable onboard audio circuitry like this. I'm sure it would be more than capable for the untrained ear.

I've never been a fan of ANY of the Reaktek offerings (Sound or NIC). I certainly fall in the "untrained ear" category, but if all else fails I do have my trusty Fiio E6!
Edited by WackyWRZ - 3/21/14 at 9:57pm
post #53 of 99
gpu shares phases with the cpu on the apu.

But really for the price MSI is chargen for this thing, it otta have a digital VRM, with mosfets as good as the gigabyte board. Hell I don't think the VRM's are the drmos mosfets they used to use on "military class 4" boards.
post #54 of 99
Oh, somehow I forgot to add that the DRAM voltage on mine... ugh! rolleyes.gif

BIOS set: 1.65V
BIOS reads: 1.640V
Test Point: 1.668V

Which while I'm remembering stuff... clock jitter is quite bad as well. Doesn't usually dip down too far below 100, but it jumps around between 99.8 and 101.5 (usually not that high but still...)

However that does bring up at least one more Pro I'll add: Some seriously fine clock adjustments that you can make. First board I've had that lets you adjust in 0.1MHz on the Baseclock! smile.gif
post #55 of 99
Monitored the CPU voltage for a bit, it fluctuates between 1.288 and 1.36 Volts while at idle, and between 1.312 and 1.352 volts while running star swarm demo in directx mode. This is at stock clocks and default settings for the CPU.

However I started playing with the bios settings, The stock setting of auto for the cpu voltage is what causes the voltage to jump around like that, change it from auto to +0.00000 and your cpu voltage otta come back into line, at idle mine floats between 1.280 to 1.288 and when running star swarm demo in dx mode the voltage floats between 1.272 and 1.288.
post #56 of 99
If you have changed any clock or multiplier settings on the CPU or NB, Quiet and cool functionality won't work unless you change it from auto to enabled.

I can't give to much of a report on overclocking until Thursday when I can replace the pos stock cooler with an h80i I have ordered.
Quote:
Only thing I have found is the reference to the Intersil ISL6377 IC which is "specifically designed" for AMD APUs.

Its an analog VRM controller, nothing fancy or remotely high end, I will reserve judgment of the mosfets until I find out exactly what they are, but I can tell you that a 4chip setup as this board has does not bod well, the "better" tech Formula keeps talking about is in 1 and 2 chip configurations, The gigabyte up4 board for example uses digital mosfets that integrate high and low side drivers and gates in a single chip. A 4chip mosfet solution is a call back to a by gone age.

That said it might still be using a doubler, until we can get a good quality picture of the numbers on the mosfets it will be hard to know.
post #57 of 99
I will be returning the board tomorrow, not only is the uefi firmware dangerously buggy, but i have found that not all of the mosfets are attached to the dragon heatsink pair ....

http://pdf1.alldatasheet.com/datasheet-pdf/view/544256/ONSEMI/NTMFS4C08NT1G.html

their are 2 of these on the bottom side of the board and they get hot as hell. Note, I advise against using your finger to test their heat!
post #58 of 99
I finally threw my 212 cooler on so I could get accurate with my testing. I set my multiplier at 43x, disabled the "Auto" voltages and set it to +0.0. Booted into Linux live CD to run stress test, and check voltages. Still seeing a 0.1v swing at load, which is better than the 0.2v swing I was seeing at idle. I still think I should be able to set a specific voltage instead of this guess and check approach with the offset. It seems like just changing the multiplier changes the "base" voltage that you're offsetting. I would much rather at least have the option to pick a voltage and roll with it. I also noticed the dragon heat-sinks were quite warm still.

On second look at my YouTube video I didn't have CPU voltage to auto there, and it was still wildly fluctuating.
Quote:
Originally Posted by luminousone11 View Post

Its an analog VRM controller, nothing fancy or remotely high end, I will reserve judgment of the mosfets until I find out exactly what they are, but I can tell you that a 4chip setup as this board has does not bod well, the "better" tech Formula keeps talking about is in 1 and 2 chip configurations, The gigabyte up4 board for example uses digital mosfets that integrate high and low side drivers and gates in a single chip. A 4chip mosfet solution is a call back to a by gone age.

I am at work, so can't verify but I am 90% sure that the manual has references to the "DigitALL" controls which is misleading. Probably a poor copy and paste knowing MSI. I wonder if these fluctuations are due to crappy hardware or poor BIOS programming. I would assume something in the BIOS has to "run" the VRM phases.
Quote:
Originally Posted by luminousone11 View Post

I will be returning the board tomorrow, not only is the uefi firmware dangerously buggy, but i have found that not all of the mosfets are attached to the dragon heatsink pair ....

I am leaning towards returning mine as well. I might hold onto it to see if MSI makes any savior BIOS updates, but I am doubting it. The same day I got this A88X is the day I finally got back my A55M board from MSI RMA (only took a month rolleyes.gif) - I think it's an omen to stay away. I also did a lot of playing with the UP4 board this weekend and it seems really solid - even the VRM sinks are actually cool to the touch. It is lacking in some of the features like the Killer NIC, Audio Boost, and it runs an older Realtek Codec, but it is also $25 cheaper ATM. I can say that the Gigabyte ET6 software is garbage, but that's what the BIOS is for.
post #59 of 99
Quote:
Originally Posted by luminousone11 View Post

I will be returning the board tomorrow, not only is the uefi firmware dangerously buggy, but i have found that not all of the mosfets are attached to the dragon heatsink pair ....

http://pdf1.alldatasheet.com/datasheet-pdf/view/544256/ONSEMI/NTMFS4C08NT1G.html

their are 2 of these on the bottom side of the board and they get hot as hell. Note, I advise against using your finger to test their heat!

I'm not exactly liking the BIOS, but I can't say I noticed there being more on the back o_0 I'll take a look right now after I post...


Mine is really cool, but I wonder if mine just is having a better contact between the Heatsink, TIM and the MOSFETs to be why? After I check the backside, I'll remove additional airflow and see if that "helps" heat mine up more.

Right now my biggest complaint is that I'm having performance issues that I suspect is memory related. Now I will admit a few things, I haven't tested at lower frequency to make sure, or plugged the other RAM back in and tried it AND my install of Win7 is very very dirty (and Pre-SP1, virtually no patches)... But overclocked to 4300MHz with DDR3-2400 and "stable" in LinX @ 6500MB test for over 30minutes (when I reached that 53C I mentioned, which is still cool compared to you guys it seems); however, when I play Fallout: New Vegas, the first time I load a save it will hang for about 3minutes before all of a sudden working. I thought it was a crash-hang and would end task, then fire up NV right away and it'd load the save no problem. Though it'd frequently do it while playing and at seemingly random times.

Even at 3700MHz with the timings loose, it did the same thing in Minecraft :\ I can't quite figure out what it might be that is doing it, since CnQ and Turbo are both disabled, so it shouldn't be some odd stalling related to that. I do still have C6 enabled but... I'm not sure if that'd fix it?

I don't know I guess it could be somehow related to my graphics card, too. I can always swap to my other 5770 or even go back to the IGP and see if it persists. In an effort to start exhausting the issues that is.... I think I will go back to the IGP first since I'll need to unplug the GFX to lift the board up anyways... Maybe this is why it took so long to get released after announcement of "availability"? I can't imagine that they made a board revision in this time and that perhaps one or both of you have the old one somehow that didn't get recalled? I got mine from NewEgg the day it was available, through their NewEgg channel (cuz it was supposed to come from Tennessee where I live, but it came from New Jersey lol). Where did you two get yours?

EDIT: Ah, yes, the two at the top under the memory phases. I would suspect that'd not be the actual MOSFETS but still no doubt would've been ideal/beneficial to be cooled. We'll see how badly my phases heat up without airflow... Idling in the BIOS my CU is only 34C (on default-failed OC settings)
Edited by Formula350 - 3/24/14 at 8:56am
post #60 of 99
The one plus I can say to their BIOS that I have really noticed compared to other UEFI BIOSes is that the mouse response is very good. Not that its a huge deal, but very noticeable. Other UEFI's seem to be a bit laggy/jumpy when it comes to mouse response. Other than that it seems like it was thrown together. I wonder if they rushed the board and or BIOS out with "bugs" to keep from having to push the release back more?

I did install a Win7 SP1 on a RAID0 array when I was playing with it. One of the first things I installed after drivers was Spotify so I could jam out during setup. I noticed stalling in installation of updates and software where the audio would start stuttering like a scratched CD - they'd last about 3 seconds at a time. This wasn't even during any gaming obviously, and I have a 760K with a 7770 and 1600 memory using XMP profiles (9-9-24), so no IGP to test. Running Prime95 while listening to music did not cause these spikes which seemed odd.

I didn't look very close at the back of my board, but I can check it out when I get home this evening for the extra mosfets. My board is a Rev 1.0 and came from Memphis, just got it Friday.

EDIT: Forgot to mention that while running P95 I also was running CPUz and HWMonitor. CPU temp was around 62 degrees and I was clocked at 4.3 GHz. I think I did have turbo and CnQ turned on, and I did notice that after about 10 seconds I would see the CPU drop down and start bouncing between 3.5/3.7 GHz. Is this what is meant by "throttling"? (Sorry if this is a dumb question - I'm still trying to learn smile.gif )
Edited by WackyWRZ - 3/24/14 at 9:07am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Motherboards
Overclock.net › Forums › AMD › AMD Motherboards › MSI A88X-G45 GAMING - FM2+ goodness