Discussion AMD Ryzen MegaThread! FAQ and Resources

Page 53 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

8350rocks

Distinguished


Exactly. If AMD felt that there was a reason to sell an 8 core processor for $1000, they would retail theirs for that.

Considering you can buy 2x 1800X for the cost of one 6900K, or 3x 1700 for that same money...I see no reason to fault any minor performance delta when you are, at worst, paying half the money...or less.
 

jaymc

Distinguished
Dec 7, 2007
614
9
18,985


Someone actually did it just can't remember who it was for the life of me...

 

jaymc

Distinguished
Dec 7, 2007
614
9
18,985



There's an update supposed to be coming out from AMD before the Ryzen 5 release on the 11th mate..
 

salgado18

Distinguished
Feb 12, 2007
966
426
19,370


I believe YT reviewers do have a lot open when they test, and possibly when they play in general. Also, note that no written review (who test in new installs, isolate the system etc) complained about smoothness. Possibly the 4c/8t Intels can't handle that many tasks at once...

... and no one noticed until then, because it looked normal. See, my first quad was an Athlon II X4 640, then I went to an FX 8120, and then to an i3-6100. None of the two AMDs give me the stuttering I get on the i3 when compiling code. Can't explain it really, the entire system chokes during high-demand tasks. And a guy said the i5 doesn't really get much better.

But no one complains, because everyone is using Intel, and they let these stuttering moments pass as normal of computing. But never once the FX-8120 let the computer hang for a milisecond. And I don't remember the Athlon X4 doing it too, although both are (and feel) slower than the i3.

I think this calls for testing in this area. I may be completely wrong here (probably, as always), but if people perceive a problem, then it's science's duty to investigate.

EDIT: like the wise guy said: ------------------------V
 
... and no one noticed until then, because it looked normal. See, my first quad was an Athlon II X4 640, then I went to an FX 8120, and then to an i3-6100. None of the two AMDs give me the stuttering I get on the i3 when compiling code. Can't explain it really, the entire system chokes during high-demand tasks. And a guy said the i5 doesn't really get much better.

You're describing a latency v bandwidth situation. Take compiling code; you're going to have a handful of very heavy threads locking out the physical CPU cores on an i3, which will lead to latency problems, even with IO, simply due to the necessary drivers being locked out for periods of time. In this case, the i3 is likely "faster" by virtue of having more powerful cores, but the FX would be more stable due to less interruption of background tasks.
 

Nope 1151

Commendable
Feb 8, 2017
70
0
1,630
Hey Juan, I saw that in your blog you reside in Spain, by any chance do you read Spanish?
<Non English language link removed by moderator>

Annnnd WBBQtech beat me to it:
http://wccftech.com/amd-ryzen-5-1600-review-leak/
 

Fizzy1

Prominent
Apr 4, 2017
3
0
510


 

Fizzy1

Prominent
Apr 4, 2017
3
0
510
They may have delayed Ryzen 5 launch so that any reviews show it in a better light whilst retailers already have them in stock ready to go
 
That video provides a lot of good data, although I don't like the set up itself used.

Still, it can be used as proof that testing with nVidia is detrimental to AMD until the drivers are optimized for Zen. Reviewers should take notice on that, just in case, when accounting all data into the conclusions.

Cheers!
 


Dubious claim, considering more powerful Ryzen CPUs don't beat the 7700k.
 

jaymc

Distinguished
Dec 7, 2007
614
9
18,985




Good point actually...

Forum is quiet people.. hope that's because were all busy writing out our questions for AMD :D

I'm looking forward to it anyway !

j
 


Just a caveat though: ironically, the 8C/16T CPUs, technically have half per-core L3 cache at their disposal for processing and more data to travel around through the fabric when thread bouncing hits (more effective threads imply more effective bounce if hit). So, for gaming purposes, at the same clock speed and with a patched scheduler or patched game that is aware of Ryzen, the L3 advantage might become noticeable. Plus all the patches and fixes AMD has put forth with the vendors.

Cheers!
 
Hmm with Ryzen's higher thread count, and good core performance, we might actually get Dev's better adopting higher core/thread usages rates for games. So people who bought those FX 8000s will finally get good use out of their mutli threaded processors. BUT I bet they'll still be screwed because they'll be stuck with slower DDR3 RAM, and also the architecture of the FX line is just not good.