egosoft.com

Oh wow! Pretty cool to hear you're seeing such a night-and-day difference!
We know from various EgoSoft statements that XR should do well on multithreading.

There is a LOT of background calculation going on in XR, that simulated economy eats cycles! I can see how having loads of CPU-power available can reduce the lag-spikes.

For me on linux, the heaviest/laggiest things I can do is flying through nebulae (e.g. Split base or Teladi space). The second-laggiest the dense asteroid zones in DeVries>Glaring Truth. Have you tried that yet?

It's not perfect, but it is very good. And I really am enjoying!

Yes, the 4.10 update did help a lot beforehand, but I still had the sluggish snail-effect all the time. Kinda funny, when watching userbenchmark's scores, that this CPU isn't really *that much* better than the i5 2500k but the in-game performance boost is noticeable.

For example now I can actually use those settings that the nVidia's GeForce experience suggests for the GFX - with the i5? No. Not by a long shot - which I still find very interesting since it certainly has the bandwidth and grunt needed.

Here is a screenshot of Windows's performance monitor actually flying in DeVries, but can't remember exactly where at.. I jumped in and went into the asteroid field with the "Ribbon" in the backround - can't remember the name of the place, and in there with somewhat combatty happenings around this is what my CPU load looked like - it is very high, but still less than everything (and with some O.C I believe I can alleviate that issue some, if I need to).

Also as stated, my GFX settings are quite high to begin with, so I am asking for a lot imo.

http://i.imgur.com/ReiNu5k.png

So 6 cores utilized - or rather 3 cores and 6 threads (can't differentiate cores and threads eh), also 7th doing part-time job. Lost the image, but the i5 2500k in the same area (same save's load point actually) had 100% load on 4 cores at all times - except if something took few clocks on itself.

Also this is what happened to my BOINC stats after upgrading from 4C/4T to 8C/16T - althought I've limited my BOINC'ing to 8threads for the time being; I got my hardware on 4th of May, installed during it and finalizing it on 5th or so abouts. And then those stats went up

http://i.imgur.com/clhcIrGh.png


{Images posted directly to the forums should not be greater than 640x480 or 100kb, oversize image now linked - Terre}

Can someone test ryzen cpu in X3AP . And write about general perfomance

^
I don't see why it would differ much... However i5 2500k vs R7 1700 differences are insignificant. Seem to me that both work just nicely.

NEMESIS165 wrote:Can someone test ryzen cpu in X3AP . And write about general performance

Don't know about X3AP performance but Jayz2cents ran Metro Last Light bench , Heaven and GTA and a Valkan api game . In March 2017
https://youtu.be/8-mMBbWHrwM?t=3m08s

start at ~ 3 min if the link doesn't load at the time .

As far as I understand it, Ryzen is still a few percentage points short of Intel on per-thread performance--Ryzen's main advantage is that you get a lot more cores for your money than you do with the Intel equivalents, but that doesn't make much difference when gaming because there really aren't many games out there which take advantage of more than 4 cores. However, if you were doing something else at the same time as gaming (e.g. streaming the game out to Twitch or whatever) then Ryzen has the advantage.

This is all before we take into account Meltdown, of course, which is going to slow things down a bit. Not massively--game benchmarks I've seen varied between 1.5% and 8% slower depending on the game--but that might be enough for AMD to retake the lead, since their chips aren't susceptible to Meltdown. However, we know that Linux at least is applying the Meltdown fixes across the board and thus slowing everybody's CPUs down, I don't know what Microsoft are planning to do.

pjknibbs wrote: However, we know that Linux at least is applying the Meltdown fixes across the board and thus slowing everybody's CPUs down, I don't know what Microsoft are planning to do.

NT and development versions of Linux do not apply the Meltdown fix to AMD CPUs - indeed, it was the AMD submission of this patch to Linux that led to people working out what the bug was pre-embargo end. It looks like most of the major distros have included the patch in their stable kernels, and it's likely to be backported in mainline too.