ASRock.com Homepage
Forum Home Forum Home > Technical Support > AMD Motherboards
  New Posts New Posts RSS Feed - AR Fatal1ty 970 Performance- RAM compatibility
  FAQ FAQ  Forum Search Search  Events   Register Register  Login Login

AR Fatal1ty 970 Performance- RAM compatibility
 Post Reply Post Reply Page  <12345 8>
Author
Message Reverse Sort Order
  Topic Search Topic Search  Topic Options Topic Options
WKjun View Drop Down
Newbie
Newbie


Joined: 17 Aug 2015
Status: Offline
Points: 56 		Post Options Post Options   Thanks (0) Thanks(0)   Quote WKjun Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2016 at 7:39am
Originally posted by PetrolHead PetrolHead wrote:

2133 MHz is a nice OC, considering the "stock" value is at best 1600 MHz with that many sticks, but I'd expect bumping the CPU-NB voltage a bit to be enough to get there. Unless it's required to get a CPU OC stable, it might not be worth the trouble though. Only a limited amount of applications outside benchmarking benefit from faster RAM and in most cases it doesn't matter whether your RAM is running @ 1333 MHz or @ 2400 MHz. If one is able to combine tighter timings with lower clock speeds, the clock speed matters even less.
AMD's "stock" value is 1866 since the FX line of CPUs. Setting it to 2133 via AMP is hardly a big accomplishment, I think. ;)
Intel stuck at 1600 for a long time, pushing DDR4 instead.
Since the CPU-NB is 2200 on all high clocked FX CPUs, there is no need to raise voltage and/or speed of it. Not until RAM frequency gets above 2200 - it has to be risen equally or higher. Some of my machines ran for some time at 2400-2500 CPU-NB. One even at 2600. With enough juice, these strange effects like graphics anomalies occured very seldom, but they happen.
Mind that this counts for Vishera cores only. Zambezi didn't like to be pushed above stock, not even 10 MHz, regardless the voltage I've tried.
Many people say, RAM frequency should be favoured to timings. On AMD (FX) a good mixture is the goal. I am very happy with my Radeon RAM kits, running at 2133 / CL10-11-11-28-1T / 1.65V.
Back to Top
wardog View Drop Down
Moderator Group
Moderator Group


Joined: 15 Jul 2015
Status: Offline
Points: 6447 		Post Options Post Options   Thanks (0) Thanks(0)   Quote wardog Quote  Post ReplyReply Direct Link To This Post Posted: 13 Jan 2016 at 2:55am
Originally posted by PetrolHead PetrolHead wrote:


Btw, the TX3 Evo you have doesn't seem to have an official recommended TDP, at least not on the manufacturer's website.


I noticed that too, odd, as most other brands do.

I based my remarks of it as it's recommended down to the foot warmer, heat the earth, arctic ice shelf be damned, Ozone? What ozone?LOL ,  Socket 754 procs. And NewEgg reviews also.

Heck, if it can keep a Socket 754 processor cool ...... I thought to myself ... I still took at peek at the NewEgg reviews.
Back to Top
PetrolHead View Drop Down
Groupie
Groupie


Joined: 07 Oct 2015
Status: Offline
Points: 403 		Post Options Post Options   Thanks (0) Thanks(0)   Quote PetrolHead Quote  Post ReplyReply Direct Link To This Post Posted: 12 Jan 2016 at 11:24pm
Originally posted by WKjun WKjun wrote:

At least my Sabertooth 990FX R2.0 runs fine with 4 x 8GB Radeon RAM @ 2133 MHz. :)
2400 is useful and usable with APUs only.


2133 MHz is a nice OC, considering the "stock" value is at best 1600 MHz with that many sticks, but I'd expect bumping the CPU-NB voltage a bit to be enough to get there. Unless it's required to get a CPU OC stable, it might not be worth the trouble though. Only a limited amount of applications outside benchmarking benefit from faster RAM and in most cases it doesn't matter whether your RAM is running @ 1333 MHz or @ 2400 MHz. If one is able to combine tighter timings with lower clock speeds, the clock speed matters even less.
Ryzen 5 1500X, ASRock AB350M Pro4, 2x8 GB G.Skill Trident Z 3466CL16, Sapphire Pulse RX Vega56 8G HBM2, Corsair RM550x, Samsung 960 EVO SSD (NVMe) 250GB, Samsung 850 EVO SSD 500 GB, Windows 10 64-bit
Back to Top
PetrolHead View Drop Down
Groupie
Groupie


Joined: 07 Oct 2015
Status: Offline
Points: 403 		Post Options Post Options   Thanks (0) Thanks(0)   Quote PetrolHead Quote  Post ReplyReply Direct Link To This Post Posted: 12 Jan 2016 at 11:08pm
Originally posted by dinin70 dinin70 wrote:

When I'm running furmark on CPU it gets pretty quickly over 60° (less than 10seconds) and I prefer stopping before harming too much the CPU.


The CPU will start throttling or will even shut down before it is damaged, assuming the safeguards work as they should. 60 C isn't even close to the danger zone of physically harming your CPU, but the CPU can become unstable well before thermal throttling kicks in.

In any case, if you didn't even have the radiator on, then something's not right. You should at the very least try installing the CPU cooler again from scratch. Remove the CPU cooler, remove all of the old TIM, apply new TIM (read some tutorials on applying the TIM beforehand), reinstall the CPU cooler and see if the temperatures remain the same.

Quote By the way, I have only Fan on the CoolerMaster: must the fan be looking upwards or downwards?


I don't really know what upwards or downwards mean in this context. If you are using the TX3 Evo - or any other tower cooler - the fan should blow the air towards the back panel. This will increase airflow over the VRM section. You should also have a case fan on the back panel sucking hot air out of the case. If you get a top-down blowing CPU cooler, then, as the term suggests, the fan should be blowing air towards the motherboard.

Btw, the TX3 Evo you have doesn't seem to have an official recommended TDP, at least not on the manufacturer's website. Based on the weight and dimensions of the heatsink, as well as the maximum rpm of the 92 mm fan, I'd say it should be okay for 125 W TDP CPUs such as the one you have. However, under full load it may struggle with an octacore and you will likely see high socket temperatures unless you can direct airflow towards the socket.
Ryzen 5 1500X, ASRock AB350M Pro4, 2x8 GB G.Skill Trident Z 3466CL16, Sapphire Pulse RX Vega56 8G HBM2, Corsair RM550x, Samsung 960 EVO SSD (NVMe) 250GB, Samsung 850 EVO SSD 500 GB, Windows 10 64-bit
Back to Top
WKjun View Drop Down
Newbie
Newbie


Joined: 17 Aug 2015
Status: Offline
Points: 56 		Post Options Post Options   Thanks (0) Thanks(0)   Quote WKjun Quote  Post ReplyReply Direct Link To This Post Posted: 11 Jan 2016 at 6:46pm
CPU-Z shows just a vague overview of some of the SPD entries. AIDA64 tells you every profile and sub-timings as well. But I can't tell, if it reads them out without a licence (Demo). Don't get confused, as some timings may not be available in your BIOS and vice versa.
 
Availability of AMD RAM Kits is really few an far between in Europe. I am lucky to have 1-2 merchants in Austria. Amazon.de and co.uk should have them too. But compare the prices, as Amazon (Marketplace) often has ridiculously high prices.
 
 
Originally posted by PetrolHead PetrolHead wrote:

That's not a huge surprise considering everything above 1866 MHz is an OC if the motherboard has more than two memory slots in total, everything above 1600 MHz is an OC if there are four slots and two or four memory modules, and everything above 1333 MHz is an OC if any of those modules are anything other than single rank.
At least my Sabertooth 990FX R2.0 runs fine with 4 x 8GB Radeon RAM @ 2133 MHz. :)
2400 is useful and usable with APUs only.
Back to Top
wardog View Drop Down
Moderator Group
Moderator Group


Joined: 15 Jul 2015
Status: Offline
Points: 6447 		Post Options Post Options   Thanks (0) Thanks(0)   Quote wardog Quote  Post ReplyReply Direct Link To This Post Posted: 09 Jan 2016 at 11:45pm
Eh, we're having translation issues. Sorry for that.

Me: "wire fan securements"
You: "safe brackets"

Not brackets, but how the fans themselves are held onto with against the actual fins of the cooler. The wire fan "clips' that Thermalright are so famous for using.

The "Evo' model is a step up from the original Hyper TX3. So better yet!
Back to Top
dinin70 View Drop Down
Newbie
Newbie


Joined: 05 Jan 2016
Status: Offline
Points: 48 		Post Options Post Options   Thanks (0) Thanks(0)   Quote dinin70 Quote  Post ReplyReply Direct Link To This Post Posted: 09 Jan 2016 at 11:26pm
Hi,

The heater probably doesn't help but I don't think it's the main reason. When I took the latest figures it was overnight when heaters are off.

When I'm running furmark on CPU it gets pretty quickly over 60° (less than 10seconds) and I prefer stopping before harming too much the CPU.

Obviously furmark is not representative of real conditions (8 cores running altogether at 100% is unlikely to happen), but yet I'm pretty sure there something wrong as you are all saying. Next week I'll be receiving the Noctua paste and I'll try to change that. I bought a manual front pannel fan controller (since I anyway need front USB because they are all broken on the Tower).

By the way, I have only Fan on the CoolerMaster: must the fan be looking upwards or downwards?

Furthermore Wardog I have the TX3 Evo (which is another product), what are those "safe brackets" you are mentionning? 

It is not fixed through "screws" but with the small  lever (if you see what I mean)



Anyway I would like to thank you already for your support. I hope the BSOD were caused by the RAM inconsistencies but yet I won't stay peace until I get that CPU cooler :)
Back to Top
wardog View Drop Down
Moderator Group
Moderator Group


Joined: 15 Jul 2015
Status: Offline
Points: 6447 		Post Options Post Options   Thanks (0) Thanks(0)   Quote wardog Quote  Post ReplyReply Direct Link To This Post Posted: 09 Jan 2016 at 11:13pm
Originally posted by PetrolHead PetrolHead wrote:


The bottom line in any case is that as long as the temperatures stay in check, it doesn't really matter which method has been used.


Couldn't agree more.

For perspective, which ever method is used, so long as the die that rests under the center of "lid' is covered with TIM on top you're good as that's the only part under the lid that makes direct contact. The outer edges of the "lid' rests atop a bead of non-heat conducting silicone.

Delidded AMD 8-Core Bulldozer Processor:
Delidded 8-core Bulldozer Module

Edited, added a couple " 's and one and for clarification purposes


Edited by wardog - 09 Jan 2016 at 11:38pm
Back to Top
PetrolHead View Drop Down
Groupie
Groupie


Joined: 07 Oct 2015
Status: Offline
Points: 403 		Post Options Post Options   Thanks (0) Thanks(0)   Quote PetrolHead Quote  Post ReplyReply Direct Link To This Post Posted: 09 Jan 2016 at 8:32pm
Originally posted by dinin70 dinin70 wrote:

Hi Petrolhead. To be honest nothing. It rebooted like 20 times since I was trying to make the RAM work (the Kingston). Finally had to place again the Corsair. So except this multiple rebooting, nothing special.
 
This morning it's turning between 39-44°
 
The PC is 15cm from a rather big Heater (Radiator).
 
In that case I think the temperatures might be a bit high and you might want to try moving the case further away from the radiator before you do anything else. The CPU temperature isn't really meaningful below 45 C and even then it's not necessarily a proper temperature due to the way AMD CPUs determine the temperature. As long as the readings in HWMonitor stay below 62 C, you should be fine when it comes to stability. If you see anything close to that without overclocking and running something like Prime95, then you probable have an issue with cooling.

The socket temperature (CPUTIN) should be reasonably reliable at least at low temperatures, and that's the one that worries me the most. If you're not doing anything much, then ideally you should be seeing temperatures nearer 30 C when idle - although maybe the FX CPU just makes the socket run a lot hotter than my Phenom II. You should check if the CPU compatibility list for your motherboard specifies the need to use a top-down blowing CPU cooler with your FX CPU. If it does, then you should find a way to provide airflow to the socket, possibly with a spot fan.

Oh, and concerning the rpm readings in HWMonitor or other programs, they may not be correct at all. However, if you've set the target temp in F-Stream to 45 C, I'd expect the fan to spin only at the speed that is needed to keep that temperature. If you're idling, then the speed may indeed be zero (and this might at least partially explain the socket temperature). I'm just guessing here, though, since I've no experience in using F-Stream. All my fan speeds are set in BIOS, and I have them on full all the time... You could try lowering the target temperature in F-Stream and seeing what happens.
Ryzen 5 1500X, ASRock AB350M Pro4, 2x8 GB G.Skill Trident Z 3466CL16, Sapphire Pulse RX Vega56 8G HBM2, Corsair RM550x, Samsung 960 EVO SSD (NVMe) 250GB, Samsung 850 EVO SSD 500 GB, Windows 10 64-bit
Back to Top
PetrolHead View Drop Down
Groupie
Groupie


Joined: 07 Oct 2015
Status: Offline
Points: 403 		Post Options Post Options   Thanks (0) Thanks(0)   Quote PetrolHead Quote  Post ReplyReply Direct Link To This Post Posted: 09 Jan 2016 at 8:14pm
Apologies for the OT to everyone.

Originally posted by wardog wardog wrote:


Count me as one of the blob folks.

Back in the old "CPUs ran hotter than hell' days I became a convert. IMO, in MY opinion and experiences, spreading it prior to fitting sometimes allows for air pockets to form in the paste/TIM between the processor and sink when the sink is secured.

While the round blob method you do need to be careful in fitting and securing, as the blob is pressed outwards no air pockets will form as it is continually coursing towards the outer periphery. Carfeful as you apply the sink that you apply even and level pressure while lowering the sink.

My 2cents anyways.


It's true that spreading it prior to fitting has the potential to form air pockets in the TIM layer. The thing about the blob method, however,  is that there's really no way of covering the whole heatspreader or the bottom of the sink without having "too much" of it. This means there will in effect be large air pockets in each corner of the heatspreader. I noticed this when I tried to follow Noctua's instructions on applying the TIM and then had to re-install the heatsink; the pea-sized TIM blob had spread to a pancake the size of a large coin, but hadn't even reached the nearest edges of the heatspreader. Of course the most important thing is to get a proper TIM layer over the die, which is a lot smaller than the heatspreader, and this is what the method had achieved. I don't know if getting a TIM layer on the rest of the heatspreader has a noticeable effect on temperatures, but my OCD says that's the way to do it. ;)

The bottom line in any case is that as long as the temperatures stay in check, it doesn't really matter which method has been used.
Ryzen 5 1500X, ASRock AB350M Pro4, 2x8 GB G.Skill Trident Z 3466CL16, Sapphire Pulse RX Vega56 8G HBM2, Corsair RM550x, Samsung 960 EVO SSD (NVMe) 250GB, Samsung 850 EVO SSD 500 GB, Windows 10 64-bit
Back to Top
 Post Reply Post Reply Page  <12345 8>
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.04
Copyright ©2001-2021 Web Wiz Ltd.

This page was generated in 0.084 seconds.