Ryzen "2" ?

[Mercutio]

Well, as someone that uses an Asus PA279 i've got no idea where your coming from on that statement ? Ok, not £800 but does me very nicely thanks.
As it happens, i use said screen with a C6H............................the C6H has had no issues with power delivery in all the time i've had it. Maybe you should just learn to clock properly, not leave things on AUTO and take control yourself ? After all, that is what clocking is about.................isn't it ?

I had.... 3 Asus branded PSU's (for my PG278Q) before getting tired of replacing them and sorted out a delta PSU with fractionally more amps/watts but the same voltage/pin. No issues since.

I getting annoyed now.

I bought the 8 pack ram to solve my instability issues and it's made things worse.

You can see my thread here.

I'm thinking of just selling up and getting a 8700k myself.

How'd you find the prime board otherwise? Saw a couple of comments here (about vdroop/similar) that have me a little worried.

 

[Vakas]

I wouldn't touch an Asus board I got the CH6 last time and had so many issues it was unbelievable.

I always use Gigabyte never had an issue.

 

[Hotwired]

This is dragging out.

It's always been the 19th, how can it possibly drag out.

 

[DragonQ]

Interesting review here comparing the R7 1700 and R7 2700X at 4 GHz on an X370 board with a beta BIOS. Seems to be a noticeable improvement in gaming just from the memory latency improvements:

Average of ~10% performance uplift in the 5 games he tested (remember: both at 4 GHz). Games still benefit from going up in RAM speeds (3200 to 3600 MHz in this case). Games that love single core performance are still way ahead on the i7-7700K (e.g. WoW) but benefit the most from running the Ryzen chip with good cooling (due to extra boost capacity).

 

[opethdisciple]

You need to reset everything to defaults, then get your memory stable at whatever speed you are aiming for and properly test it. Once that is solid start working on your CPU overclock.

I did that mate.

I set the ram to DOCP and played Origins for 4hrs.

No wobbles.

As soon as I apply my cpu overclock it falls over within 30mins.

My CPU independently is stable in Prime for 3hrs. I tested it.

It's just not happening when I apply the two together.

It even falls over at 3.7GHz now. When I've been running 3.8GHz on the cpu since September.

 

[Hotwired]

Interesting review here comparing the R7 1700 and R7 2700X

4.2 at 1.4 seems to be reasonable from this and the few other leakers.

 

[opethdisciple]

Interesting review here comparing the R7 1700 and R7 2700X at 4 GHz on an X370 board with a beta BIOS. Seems to be a noticeable improvement in gaming just from the memory latency improvements:

Average of ~10% performance uplift in the 5 games he tested (remember: both at 4 GHz). Games still benefit from going up in RAM speeds (3200 to 3600 MHz in this case). Games that love single core performance are still way ahead on the i7-7700K (e.g. WoW) but benefit the most from running the Ryzen chip with good cooling (due to extra boost capacity).

So memory support is down to the chip and not the mobo?

So for example if I got the 2700x and put 3600MHz ram in it it shouldn't be limited by the mobo?

 

[DragonQ]

So memory support is down to the chip and not the mobo?

So for example if I got the 2700x and put 3600MHz ram in it it shouldn't be limited by the mobo?

Theoretically yes, the IMC is integrated in the CPU after all (and this has been the case for Intel and AMD CPUs for at least a decade).

 

[opethdisciple]

Theoretically yes, the IMC is integrated in the CPU after all (and this has been the case for Intel and AMD CPUs for at least a decade).

I guess all will be revealed on Thu.

I'm in no man's land at the moment. I bought this 8 pack ram to replace my Corsair which I have agreed to sell today. It turns out I cannot run the 8 pack ram any faster than my Corsair after all.

I'm starting to feel skeptical if the 2700x would fair any better when it comes to overclocking.

I also fear being hobbled by my x370 board.

Tempted to sell up and by a 8700k.

 

[4K8KW10]

I'm starting to feel skeptical if the 2700x would fair any better when it comes to overclocking.

I also fear being hobbled by my x370 board.

Tempted to sell up and by a 8700k.

You don't need that turd Just stay on AM4 and next year you will get a very significant core count and performance upgrade.

 

[opethdisciple]

You don't need that turd Just stay on AM4 and next year you will get a very significant core count and performance upgrade.

Yea that's true.

 

[Sao_Sin]

You don't need that turd Just stay on AM4 and next year you will get a very significant core count and performance upgrade.

Surely Intel's next gen release is going to be looking to knock it out of the park too due to all the recent bad press?

 

[orbitalwalsh]

this is prob been posted up earlier so will spoiler it

Spoiler: z490

thought it was a bit of a question mark if AMD would actually muscle in on Intel and all it Z ....

 

[DragonQ]

Surely Intel's next gen release is going to be looking to knock it out of the park too due to all the recent bad press?

You'd expect them to be spectre and meltdown "proof" out of the box, but that could simply mean the default microcode is updated. I don't think they'll be able to completely fix the spectre issues without a fairly significant redesign though, and they haven't done that for a long, long time. Do not expect it in their next release.

 

[opethdisciple]

So the question is...

This 8 Pack 3200Mhz ram I bought, if I upgrade to the 2700x should I send this ram back and get 3600MHz or just keep it?

I've had it for 6 days so need to make a decision quick.

 

[mmj_uk]

Is anyone else really confused about voltages these days? it used to be the case that the smaller the die process the more dangerous voltage was because of electromigration. When Ryzen first came out AMD recommended 1.365V as a maximum but now Ryzen+ is out on an even smaller process and it's running at 1.4V+ stock with boost? Most 12-16nm parts today run at 1.0-1.2V and 1.4V was considered high in the 32nm era. What changed?

 

[opethdisciple]

Is anyone else really confused about voltages these days? it used to be the case that the smaller the die process the more dangerous voltage was because of electromigration. When Ryzen first came out AMD recommended 1.365V as a maximum but now Ryzen+ is out on an even smaller process and it's running at 1.4V+ stock with boost? Most 12-16nm parts today run at 1.0-1.2V and 1.4V was considered high in the 32nm era. What changed?

I think up too 1.4v was considered fine for daily. That's what that Robert Hallock said in one of his videos.

 

[DragonQ]

Global Foundries' 14 nm process was optimised for low power applications, their 12 nm is probably more balanced when it comes to performance v power, hence higher voltages.

 

[Zeed]

@Mercutio buying Branded products is false logic.
If i want psu i wobt buy asus i go for seasonic or superflowe maybe corsair
I want ram wont buy asus but gskill teamgroup

Like if i want ro buy running trainers i wont buy damn ferrari trainers cause what 2 **** do they know about running?? I run in Mizunos cause they been in trainers for years.

Branded products that are made by other company with just logo slammed on is for suckers.

One always goes for source designer / producent.. O ye and I'm avoiding razers chinese junk lol

Feom asus id only buy Motherboard and Monitor everything else id get from better options.

Dj dree beats anyone ?? Lol

 

[Overlag]

So memory support is down to the chip and not the mobo?

So for example if I got the 2700x and put 3600MHz ram in it it shouldn't be limited by the mobo?

its a bit of both. The MB has got to be able to supply stable voltage at those frequencies and also the traces are optimised for certain specs, that have probably changed with the Ryzen2 series.