AMD Ryzen vs 7700k

This is interesting, so lets bump this.

I knew some of you already watched this

>>59877999
Delete this. Now. Before we come to blows.

>>59877999
Frametimes literally don't matter. All that matters is the static FPS. I bet you don't even know what goes in to making an average an average, stupid newfags.

Can't you just keep fanboying out of this and discuss an interesting video?

>>59877999
ONLY AMD ALLOWED REVIEWERS AND NOBODY ELSE REEEEEEEEEEEEEEEEE

>>59878357
>stupid newfags
>can't even bait properly
man I'll tell ya

>>59877999
Can't watch the video right now, but do the same spikes happen on Intel 6 and 8 core? Do they happen on 4 core Ryzen?
Would be interesting to see whether these issues stem from quad cores reaching their limit.

>>59878407
The guy's main rig is dual xeons, the guy doesn't even care for gaming, he told his cronies to play some games while he launched the profiler tools and used some common sense to confirm and interpret the results in more detail than a regular monkey at Gamers Nexus or LTT does.

You're literally foaming at the mouth

>>59877999
wendell strikes a nice point in this. there 20000 frames which they analyzed, and only 4 of them had this shitty frame times.

and this hiccups were only repeatable in GTA V. even in a mess like fallout 4 they couldn't repeat this kind of behaviour.

and none of them can deduce which system is which one while playing.
basically it is all for nothing.

>>59878425
only 7700k vs 1800x.

>>59878425
He only tested a 7700k for this, but in the last video 2 days ago when they were doing the blind tests (which led to the OP's video) the 6 core Intel exhibited the same behavior

It's probably some kind of I/O spike, or just some pipeline pitfall, can't really say.

>>59877999
>>59878357
Just another AMDrone falling the frametime marketing meme. Move along.

>>59878438
If they were hitting some 160FPS on this, how much minutes does it take to reach 20000 frames? Some two minutes?
If this kept happening several times per minute, it would be pretty annoying.

>>59878438
Fallout 4 is nuts though, you can very different results with the same hardware in the same areas doing the same thing, it's a nonsense engine.

>>59878438
Repeatable means you can repeat it on demand, it does not give you any specification as to how often it happens.

>>59878497
yes, two minutes and 5 seconds.
>>59878564
yes I know, that's why I used it??

>>59878497
Even more, the average was over 170 for the 7700k

Hopefully for Intel this is a game specific problem, if this can be repeated in different games, that's pretty awful.

File: Unbenannt1.jpg (57KB, 652x486px) Image search: [Google]

57KB, 652x486px

Well... uhh yeah.

I'm guessing frametime graphs are better than 1% and 0.1%

>>59878699
Seriously why would ANYONE fall for the QUAD CORE MEME in THE CURRENT YEAR?

>>59877999
This should result in some interesting implication for future testing, how can you test this now? Time spend over a certain amount of ms?

>>59878745
>Time spend over a certain amount of ms?
frame times
this picture >>59878699

>>59878759
If the graph is long enough or low resolution or just shittily colored these sporadic high frame times that are a single freezing frame are hard to spot.

>>59878786
if a big scale would be used it can be seen, as wendell does in the video.

>>59878585
>yes I know, that's why I used it??
No, you used it wrong.
You assumed fallout 4 didn't have those hiccups, but they did, they just couldn't repeat them on demand.

>>59877999
i've had this issue with my 4670k in gtav

Since this was also seen on the 6 core Intels this has more than likely got nothing to with cores.

A not very commented part of Ryzen is the ability to run every core, even over the CCX and MCM and sockets, at the same execution latency as other cores(I guess there's either a timer dedicated to this, there's some predictable command unit or it's built into the infinity fabric, I think it's due to IF because it scales to MCM and over sockets) this ensures complete consistency while Intel systems might have synchronization spikes at times.

>>59878859
I definitely meant fallout 4 have hiccups, but in a given circuit, they couldn't repeat the hiccups at the same places like GTA V. you just didn't understand.

>>59877999
except frametimes were already tested before, by multiple outlets and it always showed ryzen as worse than intel.

http://techreport.com/review/31366/amd-ryzen-7-1800x-ryzen-7-1700x-and-ryzen-7-1700-cpus-reviewed/5

>>59877999
This little video is so much more detailed and worth than most of the youtube and reviewer garbage who just shit out number and are done with it with almost no curiosity or investigation.

Bonus points for Wendell being an actual knowledgeable guy about the workings of hardware and software.

>>59879004
wendell says the same thing. 7700k overall has better frame times.
there is 4 hiccups in 20.000 with intel. ryzen didn't have that.

really not a big deal.

>>59879004
>day 1 review
>only 2 games
>doesn't really understand the point of the damn thread but still comes in gifting us his godly knowledge