• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** AMD "Zen" thread (inc AM4/APU discussion) ***

You're right, except spending money on a new system or part new system and people also look ahead. With skylake x 6-8 months away, a lot of people will hold onto their cash for now. They want to know which platform to pick for now and the future which will be heavily influenced by current offerings and price. If those 6 core skylake chips clock to around 4.5 and these AMD optimisation's come into play fully by then, it should make for an interesting choice given Intel will be more costly.

We already had those chips. The 5820K offers that pretty much. If Skylake X offers a unified socket with 5 years of support and a 20 IPC uplift over Skylake it will be interesting. I can't see that happening though. Chances are Skylake X will be 2% faster than Broadwell-E for 20% more money on yet another dead end socket.
 
For anyone who would be interested in the power usage of a Ryzen system here are my results (PC unit only):

System and connected peripherals:
1700 @ 3.6Ghz (All cores) on 1.112volts
MSI B350 Tomahawk
16GB Crucial 2400Mhz RAM (2x8)
Crucial MX300 750GB SSD
3TB WD Red HDD
EVGA FTW 1070 8GB
Sennheiser PC 363D headset
Corsair K65 RGB Keyboard
Logitech G303

Pushing a 3440x1440 monitor


m6fkAib.jpg
89MZjkr.jpg
zBJb8hg.jpg
vXwS84D.jpg
36arVw2.jpg
zcghpOi.jpg

I am very impressed considering it's a 8C16T monster!!!


Thx Scramz
 

This video proves there is something wrong in some major games when all cores + SMT are enabled. Some of the latest games run noticeably better in the 4c4t mode.
Not sure why AMD dismissed this issue.

edit: conspiracy theory mode. They want to make their R5 Ryzen the main gaming chip so in their interest for the R7 to perform sub-par in games:eek:.
Yep. 4c/4t on a single CCX causes all 4 cores to be pegged at 100%, yet the FPS is significantly higher than when using 8c/16t mode. Something is wrong.

Would like to see the same test on Windows 7.

Hmm so these guys getting better FPS on Windows 10 compared to Windows 7 by disabling core parking. Interesting that on Windows 7, only the virtual SMT cores are parked, which basically means no cores are parked, whereas on Windows 10 cores are actually parked, hurting performance.

Windows 10 + HPET off + High Performance power profile + core parking disabled gives them the best performance figures.
 
Last edited:
We already had those chips. The 5820K offers that pretty much. If Skylake X offers a unified socket with 5 years of support and a 20 IPC uplift over Skylake it will be interesting. I can't see that happening though. Chances are Skylake X will be 2% faster than Broadwell-E for 20% more money on yet another dead end socket.
Which is why I moved away from intel as soon as AMD had something that was competitive, intel just want to bleed you dry every year for no more than a 5% gain
 
We already had those chips. The 5820K offers that pretty much. If Skylake X offers a unified socket with 5 years of support and a 20 IPC uplift over Skylake it will be interesting. I can't see that happening though. Chances are Skylake X will be 2% faster than Broadwell-E for 20% more money on yet another dead end socket.

It's unlikely to be huge but 2%... Quite a few games should see a nice boost. All guess work as it stands regardless.
 
Why does it run better with HPET off? AMD created HPET lol.

Because the implimentation in W10 without CPU drivers specific for Ryzen to pick up HPET properly it is a hinderence that should have been resolved prior to release. Of course at this time W10 has very good Intel drivers for their CPU's that are utilising HPET correctly.

It doesn't really matter who invested the system if it is only tuned for one vendor properly by Microsoft. I just don't understand what AMD have been testing with W10 and why they couldn't see all these issues when they clearly had Intel systems to the dozen to compare principles too.
 
Back
Top Bottom