Crashing while in game on a new system (some advice please guys)

[switcheHHH]

If you're heatsink is redders but your cooling rad is barely warm then that would indicate a flow problem. If they are similar temperatures then it would indicate the heat isn't being transferred from the cpu effectively.

Or the cpu isn't really getting that hot ! lol

1. I'm not overclocking so will it make THAT much of a difference with a fan pointing at the VRM's?

2. no I don't want to buy anything else if I can help it but it's looking bleak

3. well the grill was 'warm' at best so i don't know what that tells you...

 

[switcheHHH]

your CPU is effectively factory overclocked, it will be drawing quite a bit more power compared with the CPU's the motherboard was designed to deal with

Point taken, I have a fan so I shall put that on while reseating the cooler.

 

[switcheHHH]

1. I'm not overclocking so will it make THAT much of a difference with a fan pointing at the VRM's?

2. no I don't want to buy anything else if I can help it but it's looking bleak

3. well the grill was 'warm' at best so i don't know what that tells you...

Effectively you are overclocking it's just you've paid for it by the purchase of your cpu. If you bought an fx8320 you would end up tweaking the bios to a similar end result as the fx9590 is at stock.
When using a liquid cooler, adding a cooling fan or 2 over the socket and vrm area can always help, it depends on how much voltage you are pushing.
1.46v even on your board is not pushing it so I don't get why people feel it is.


If the rad was warm but the heatsink was red hot then

Ok so BIOS is up to date, chipset is current. I reseated the cooler made sure it was firmly on (it is) the pump works everything seems to work.

I did a stress test, I crashed out at this temp.... lol. It can't be the cooler, not when it crashed out at those temps, I was going higher than that and not crashing before. (see previous screenshots of stress tests)
Someone save me from this hell...

 

[switcheHHH]

what bios are you using?

That one which is the latest (2501), which was already running (I checked my current BIOS version) so I didn't update anything as it was already done.

ALSO my cooler makes a squirting type sound every so odten? Always has done and it never gave me problems on my old setup, wanted to check if that was normal?

=============================================================================================

Insane Idea

I have two PSU's....what if I used both? use my 750W for CPU and HDD's / 520W for my graphics card.

It's only feeding power and with the extra 520W I can see if the PSU really is the issue.....right?

I don't really have many (if any) options left. thoughts?

 

[switcheHHH]

Package temps still high and it sucks we don't have vrm temps.
Turn off cpb in the bios and set the CPU multiplier to 22.
Then next test get the biggest fan you can find to aim over the vrm heatsink.
Let's see if the temps improve.

What about my crazy idea?
Okay, will give it a go.
Have you got skype/steam/some kind of chat platform I could add you on? Seeing as you reply the most it would make life easier I think.

 

[switcheHHH]

Afraid I don't use any chat apps, last one I used was MSN many years ago

Fair enough pal. Okay well I'll try what you said above, brb.

 

[switcheHHH]

Jut quickly, how do i reset BIOS, is it just ''load optismised defaults''?

As I hit OC tuner by mistake an it probably OC'ed everything, crashed 3 times just booting up

 

[switcheHHH]

Yeah load optimised should do it, or clear the cmos

I did what you asked, restarted and instant crash... so i had to load defaults again.

CPU temp is massive....

 

[switcheHHH]

That's crazy high temp for 1.40v
I'm just not sure what to suggest, is it the cooler, is it the motherboard, is it the psu, is it the CPU?
If you can reduce the CPU voltage to 1.35v turn off Turbo core and lower the CPU multi to 20.
Then it'll run at 4ghz at 1.35v, then we can see what's going on

Okay. that's done (see below), I also loaded HW Monitor while idle.

I will do a stress test now and post that too

 

[switcheHHH]

Crashed within 2 seconds of stress testing.....

 

[switcheHHH]

When it crashed it def was running at 4ghz?

yep

 

[switcheHHH]

OK in digi power control set your CPU load line calibration to medium

Will do it when I get hone, this is proving to be a nightmare lol

 

[switcheHHH]

Ok lets get the point here, it looks like a multitude of issues (probably all my fault for not looking into it enough)

Look at these reviews
https://www.overclockers.co.uk/showproduct.php?prodid=CP-360-AM&tool=3

my main concern is my motherboard cannot handle this CPU properly at stock. I know some people have said my main board is fine however....
''this CPU needs 220W TDP''.....my main board does NOT provide this, as previously pointed out by -Marine-RX179- , so would this not be the main cause of my issue?
If not please elaborate.

Also another review stating I need a PSU 800W or above.. I'm rocking a 750W so there is anything potential problem.

I think the only reason the temps are so high is because of the amount of work I am making all my components do.

I could be wrong but that is honestly how it seems right now.

 

[switcheHHH]

One of those reviews is mine I think (4 stars 'cos nothing is perfect).

I have a seasonic x-650 gold PSU. I have a Radeon 7870XT GPU (185w), 16GB of mixed ram, 3 x SSD and the 9590.

My overclocked 8350 (4.7) ran at about the same wattage as the 9590 (4.7) does when at full load. Also only a little more than my mates overclocked i7-2600 at load.

My 9590 runs at 0.08V less than the 8350 did!

I have seen a maximum of 450watts (at the wall) using prime 95 (CPU) and heaven benchmark (GPU) concurrently on the 8350, I believe that the 9590 would not be vastly different.

I think that it is the settings that need refining. However the option remains I suppose to sell the 9590 and get an 83** instead.

That confirms it is less likely to be my power unit.

That doesn't explain why undervolting your cpu and underclocking your cpu still shows masdive spikes in package temps. Unless the board is faulty it should easily manage that 4ghz at 1.35v. Maybe you'd be best returning the board

The cooler worked on my other setup fine. there is no way it can be that to be honest.

What options do I have if I go for a different main board?

 

[switcheHHH]

Gigabyte 990xa ud3,£80
Gigabyte 990fxa ud3
Asus sabertooth

And a fx8320/e or your fx9590 if you keep it

The Asus Sabertooth says
- Supports CPU up to 8 cores
- Supports CPU up to 140 W <---isn't this the TDP? if so then its not enough

Actually looking at your latest screenshot, I think there's definitely some problem with the cooling, as there's no way the temp of the CPU should be that high when in bios.

I think there's definitely something need to be check about your H90. I own the H105 myself, so I have some idea on how the H90 should work as well.

Theses AIO units are not difficult to mount at all, since securing them up only require tightening the 4 screws, so contact wise shouldn't be an issue I would assume. Moving on, there's the pump and the fans...

1. The pump H90 is powered by the 3pin fan header like connector, plugging onto the motherboard- fan header on the motherboard which which the pump is connector to...have you made sure that particular header is not under Q-fan speed control? Asus's otherboard's Q-fan speed control feature which you can select different fan speed profile (which use different level of voltage) to control the speed of the fans connected to Chasis fans's fan-header on the motherboard. If you are powering the pump from the fan-header, you have to make sure the Q-fan feature is disable so it would provide full power to the pump, rather than reduced voltage.

2. The fans of the H90- where are they connected to on the motherboard? The CPU fan-headers? Like the Chasis Fans, the CPU fan-headers would have Q-fan feature as well...and to make sure the fans running on full speed with the features disabled, or at least being ran on "turbo" profile rather than at the lower fan speed on "standard" or "silent".

I will look into this further tonight, I am taking my system round to a friend who works in a custom store (so he knows his stuff) I will refer everything in this thread to him as well and hopefully I can just move on with getting this sorted.
He is a last resort before I start crying to OC about needing to change stuff lol..

 

[switcheHHH]

Update

So I can confirm the following..

It is not the graphics card causing the issue, tested my 960 and a couple other cards, no change in the issues I was having.

The cooler works 100%, there is NO problem with this what so ever. tested several coolers and got the same problem, also used my current cooler on other high end spec builds just to be sure.

I tried some different RAM as well just for the hell of it, obviously no change when I tested.

Doesn't look like it's the PSU either.

That being said it's now either the CPU or the main board. I will RMA the main board and go for a better board that is more suited for the processor.

Thanks for all your help guys, it's been appreciated but it does look like it's the board that is causing the issue (despite the temperatures we were seeing) problem with computers is that every issue if different even if symptoms are the same.

I will update next week when it comes in and I have tested again.

 

[switcheHHH]

Glad to hear you have a better idea about what's what's...after you have taken your system your your friend (?) who know more about PC.

Hope your problem will resolve once the motherboard has been replaced.

Not saying he knows more than people here, but he knows more than me
Also was able to test everything properly as there were other parts available to change out enabling me to narrow down the cause of the issue.

Yeah I hope it gets resolved too! *fingers crossed*

 

[switcheHHH]

Glad you managed to rule out components by swapping parts.
What board you looking at for round 2 ?

Probably one of the Gigabyte boards you recommended

 

[switcheHHH]

I highly recommend the 990fxa ud3 rev 4.
http://www.overclockers.co.uk/showproduct.php?prodid=MB-350-GI

Here's a really good review showing the potential of the board with an fx9590 and overclocking it, and explaining the vrm phase section.

http://www.modders-inc.com/gigabyte-990fxa-ud3-rev-4-0-motherboard-review/4/


vrm and motherboard overview
http://www.modders-inc.com/gigabyte-990fxa-ud3-rev-4-0-motherboard-review/2/

Is that 100% the rev 4.0? I can see 3.0 mentioned in the product description.
But thanks I will probably get that one if it is rev 4.0.

Good choice (and interesting review, I had not seen it before) make sure it is rev. 4 though.

I noticed in the other thread that you were buying additional fans, would be useful as airflow can make a huge difference.

good luck, andy.

Yeah I am going to get an extra 2 thank you!

If they're selling a rev3 then politely tell the seller to.put it in their nearest bin.
The rev4 was gigabytes quiet fix without admitting the rev3 was broken.

If the one above from OC isn't the rev4 I 'll go with a different board.

Hope I'm wrong but I suspect the problem is actually the CPU due to the package temps being completely out of whack.

Hopefully all goes well for you and the rma goes easily. Good luck.

It's possible it's the CPU, but while my board doesn't really support the CPU I need to go with that. I might just RMA the CPU too with the board an get them to test it.

 

[switcheHHH]

UPDATE

SO I have ordered the Sabertooth motherboard
https://www.overclockers.co.uk/showproduct.php?prodid=MB-576-AS

I have also ordered this thermal paste
https://www.overclockers.co.uk/showproduct.php?prodid=TH-003-AR

Although it shouldn't be the CPU or the motherboard, it's all I have left to test... =/

I will let you know how I get on with this stuff after work tomorrow!