Performance. FPS. Acceleration. +Odyssey (Assassin's Creed: Origins)

Keep us informed

An interesting test

VOVAN WOLF
Assassins are made with emphasis on top-iron since 2012,if not sooner.Yubiki assassins know how to overload the system,especially the CPU.On my 7950 Benchmark in Origin somehow better passes than the game itself.
I now pass the AC3.Game of 2012,she's risen to 87% in new York manages to ship in a pair of 570GTX:
Boston smokes nervously on the sidelines

[Denis]
Oh now disgraced:Top iron doesn't. Well 2700 and 80% load at 80 frames. There loads 16 threads / 8 cores? The same and 1600S, in the coupe it is more powerful than 4790k, but how the game manages to ship under 100, is not clear. It feels like the article is real, the more powerful the iron, the more download.
And here's another:I'm a lousy scholar the Mat. part, I am ashamed, BUT:
Conventionally, the graphics card is constantly at 90-100% loaded (usually 95-98%), but the cost was through Opera, and percents as numbers grew. I always thought that the focus download graphics means its disclosure (disgraceful term). Ie Yes, FPS is the result of effective work of the CPU with the game engine, BUT if the graphics card is 99%, what then? And then we have to understand that the work performed by the processor associated with the graphics part. Plus, the more powerful, the more work useless throws the game engine.
It is still in the article was: https://www.playground.ru/blogs/assassins_creed_odyssey/modder_vyyavil_prichinu_plohoj_proizvoditelnost_odyssey-319867/
1 and 2
It was also noted that the developers focus on the console, and do not consider a high performance personal computer.
the game engine loads, then unloads, and later restarts the same assets of hundreds of thousands of times within an hour, while the video memory of the GTX 1080 Ti loaded with only 25%

The feeling, the game is looking at a more powerful processor stupidly throws an empty load on it.

VOVAN WOLF
VOVAN WOLF wrote:
I always thought that the focus download graphics means its disclosure (disgraceful term)
I thought so too,until I saw the AC3 test on FX-8350 + 960GTX:
Spoiler
And i7 2600k+960GTX:
Spoiler
Here are my Originals:
Spoiler
The game on the same setting can't

VOVAN WOLF wrote:
It feels like the article is real, the more powerful the iron, the more download.
And back again to the old Assassinio:
Spoiler
The dryer was loaded about the same as Riesen.He's the only skinny for AC3.
While the map does not change at least for 1060GTX,the Origin will not stutter more

VOVAN WOLF wrote:
Oh the shame and disgrace now
Never think that.You spread your observations.

VOVAN WOLF wrote:
Top iron doesn't.
Yubiki make it so that it solves.Accessor would run on any core i5 4670 or 3570К,or 2500К

Vova,you could used to test the Origin on 1060 in FHD on high settings? If Alexandria had survived,then,preferably in it,do not benchmark and that game. And shoot VIDOS Shadowplay.If it is not difficult,of course.I laid eyes on 1060.But buy or not,not sure yet.

[Denis]
[Denis] wrote:
you could used to test the Origin on 1060 in FHD on high settings?
Yes, already done. Well, almost. Yesterday I switched from 1600S + 1060 6Gb at those settings, higher owczary lying, there was cloud vklyucheny, he gave 55 frames.
VIDOS record, I have somewhere a couple of hours. Will do, I'll post it on the channel, unsubscribe.

VOVAN WOLF
Thank You, Vova.You're a good man)

[Denis]
Suffer with converting and uploading them for the first time did, thought easier.
Spoiler
Correct, should - be remake. Something extra was thrown on the screen.

VOVAN WOLF
Vova handsome,high FHD 1060 just enough.Why two CPU?

[Denis]
Oh, I don't remember, it's the old sets. Riesen noted at the TWO temperatures. First - Tdie, indicates the temperature of the crystal, and the second Tctl = Tdie+20 degrees. I have not found how to change afterburner and real, at least early. Because the second indicator is hwinfo64 c marked at the same time the voltage and frequency. Usually, just after RAM these numbers are not displayed, because hwinfo64 most of the time closed.

VOVAN WOLF
Anyway thank you very much.You gave me hope in the opportunity to play one of the most gluttonous games).

[Denis]
Full:
You're welcome. Contact.

VOVAN WOLF
Fucking CE,Riesen 100% load.
Someone bombed or something from your theme?Minus

Yes uzbagoytes, amuda just made shit cent, which is not all and not all games work fine, the engine is the last of the assassins proof, let's see how the Intel resting in the background of Ryazan

VOVAN WOLF
Here ran on the ultra review on 115 Locke 60 fps in Alexandria (the most difficult locations)

In the end a stable 60 with the start of the game after rebooting the computer, i.e. on cold, and then there are some on this occasion that they were shouting (it showed the Windows in the record that there was no question that the game's first run, a minute and a half left to enter the password in the input Windows, start yupley and steam, the incentive to enter the password code Guarda - pin).
As a result, even 0.1% below 58 the minimum has not fallen and it is based on battles in the game + the game record. Now show me something like that in the risen 2700Х )))
The game runs perfectly smoothly and if before the old percent some games, too, was stable under 60 in any team fights the fps can SAG and now everything is just constantly stable.

VOVAN WOLF
You just made FAKE and all the brazen lied, optimizer you. If you can't do anything then why make false quiz to raise its CDA???
Here the data of the bench you have exposed in your ability to there any optimization ))))))))) the second skin is CLEARLY visible that the average rendering frame, the GPU 11мс and in the first 14. And if the frequency is the same in the second case, just reduced some times the graphics card is working faster and all that. Also monitoring shows that in the first case as VRAM and RAM are involved is more and hence the settings above. So there is nothing to cheat the optimizer you are our )))))))))))
You even promised to upload a video in another thread to refute but merged tail between his legs ))))))))))) deleting all the posts )))))))))))))))

KillKammbest
Recorded video:
Spoiler
Yesterday when running the tests passed fine, after starting the torrent and idle night, began MicroProse. It was a surprise. I remembered a glitch with dozens of microprism and the problem of cache memory. Launched MapReduce, clogged the system cache was 50%, i.e. 450 MB of the 800 there is nothing. Cleaned and MicroProse passed! It is not necessary to believe it, You PC once a week for an hour run, and I have the clock working, it makes you wonder. Had to install a program to clean.
It's funny that now the system cache available at 134 MB, and employs 78%. Although, before it was over. Still need to figure out.
Spoiler
By the way, how's the video quality? I think it's very good. That's 59 frame no. Went to drive the processor and the RAM.
4790k vs 7700k
It shows 2% more without recording, ie 520. And it is against 7700k with Boston 4500 and tehprotsessom 14нм..

VOVAN WOLF
Aren't you the one screaming that I have in the video, everything slows down what made the dumb conclusion that I have in the game everything slows down??? At the very same video even worse than me ))) I'm talking about the Jumpy picture.
And why should I VIDOS test, it even i5 with no issues at stable 60 will take out, you show how you have in Alexandria runs with a recording from the desktop and even then you in the test before entry ran a test to the in-memory ))) who's muhlevich ))))))))
Yes, and even looked up the settings, you're not ULTRA and below??? where is ultra??? vidyuhi their exports without problems, I have even 2K on the ultra stable 60!!! with just a few small failures.
Here is the girder, and with the entry after reboot from the desktop.
Spoiler
VOVAN WOLF wrote:
You PC once a week for an hour run, and I have the clock working, it makes you wonder. Had to install a program to clean.
I was just wondering what the problem is to restart the computer??? this is something impossible, incredible, super challenging???
So let's VIDOS after a reboot (because you're not even in the test did a run-through before recording) with the recording from the desktop and showing the TIME FRAME which reflects the intervals of output frames, with a run in Alexandria and for the load with the launch of the eagle. But I think you're just going to merge because you just wouldn't be the same stability and not to shame faded )))
I want to see how HT will affect the stability time frame, and drawdown at 1% and 0.1%.

KillKammbest
KillKammbest wrote:
At the very same video even worse than me ))) I'm talking about the Jumpy picture.
Off vertical sincronizate. I noticed that, but as we scored, and then looked closer, in General, the inclusion of the rifle and solved the issue (was Lok using RTTS).
KillKammbest wrote:
and then you even test before recording ran a test to the in-memory )))
Clear figure You these numbers are constantly look, as all do. To loaded, the objectivity of the results, that was not the test HDD.
KillKammbest wrote:
Yes, and even looked up the settings, you're not ULTRA and below??? where is ultra???
Left standard with 4790k. Did not change them.
KillKammbest wrote:
I have even 2K on the ultra stable 60!!!
I'm very happy for Your 2048х1080, but I have a little less than 1920x1080 and the resolution modifier is 120%. Me just enough. Besides Benedicam Otel has 10 frames, Desk to bandika turned out well, but the game sucks recorded.Now I'm back with shadow play.
In short here is the video without the rifle, so that the strip will be.
Spoiler
I ran three times, just because of course reloading, here is the video from last time is the drop in frame 55 at the end. Two out of three times 60 / 60.
KillKammbest wrote:
I was just wondering what the problem is to restart the computer??? this is something impossible, incredible, super challenging???
What agressing this, no to PC Boyar, brother to sympathize, they say Windows is complete garbage, there should have been the questions what to ask.
KillKammbest wrote:
you just wouldn't be the same stability and not to shame faded )))
This has never happened before.
KillKammbest wrote:
the run in Alexandria for loading with the launch of the eagle. I want to see how HT will affect the stability time frame, and drawdown at 1% and 0.1%.
In short, there is only the bench, it is easier. In Alexandria more difficult, the fact is that while I progrules objects percent were under 90-95% loaded, was drops. Settled, stable 60 was startled, of course got in a fight, drove the eagle. I have another conflict was rifle with RTTS and Locke watched.
Finding one percents does not pull, needs acceleration. Then I experimented, and I can not pass Prime 95 with AVX instructions, or have something wrong configured or done. Hangs everything.
Spoiler
Video later is possible, as with the processor will understand.
Spoiler


So questions:
Adaptiv is mod 1.34 (know a lot) +/- 0,02 offset. 4700 was the dispersal (-2) CPU Ratio Offset When Running AVX, i.e. the calculation of the 4500 (as in the Origin will work). Fails Prime 95 in any crashes. Voltage too small or have 8+4 pin on the Board to connect.
As You have, what performance, acceleration, voltage and work instructions.