• 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.

Intel to launch 6 core Coffee Lake-S CPUs & Z370 chipset 5 October 2017

CAT-THE-FIFTH said:
Total side-note. Any of you with Coffee Lake rigs seen any noticeable improvements with the Fall Creators update?? I know Forza7 stuttering has been fixed,but not seen many Coffee Lake owners chime in with the various threads on the subject.
Click to expand...

The update completely broke my install. Went from a rock solid system to constant SYSTEM_SERVICE_EXCEPTION blue screens and broken home directory permissions.

Managed to stabilise the system messing around in safe mode, but then in games I had frequent major FPS drops whereas before the update every game was like butter.

Ended up having to nuke my windows install, which fixed it, and I haven't updated to the fall update again yet...
 
edscdk said:
It's been mentioned loads of time on reviews that it's impossible to compare multi player games as every test would be different..
Click to expand...
True, every test would be different but if you run a lengthy test, and do it multiple times, a fairly consistent average would form I think as long as you made it as reproducible as possible (ie, same number of players, same length of time). Single player would probably be better but same conditions (length of time etc)
The problem with a short benchmark and a single value for max, min and average FPS is that you could have two systems with the exact same values that actually play entirely differently.
 
Prozn said:
The update completely broke my install. Went from a rock solid system to constant SYSTEM_SERVICE_EXCEPTION blue screens and broken home directory permissions.

Managed to stabilise the system messing around in safe mode, but then in games I had frequent major FPS drops whereas before the update every game was like butter.

Ended up having to nuke my windows install, which fixed it, and I haven't updated to the fall update again yet...
Click to expand...

That sucks! :(
 
gavinh87 said:
x2 computers next to each other, same server and just follow each other around.
Don't see whats so difficult about that tbh.
Click to expand...
But what would be the purpose of it? Who in the real would would run a game like that, they would be different every time for every user.

Darren has the better idea just run it 3 times and work out an average.
 
Doobedoo said:
But what would be the purpose of it? Who in the real would would run a game like that, they would be different every time for every user.

Darren has the better idea just run it 3 times and work out an average.
Click to expand...

CPU1/GPU1 in the same game at the same time and compare against CPU2/GPU2.
What would NOT be the point of that? It would clearly show the real time differences.
 
I think Gavin87 is saying two computers with each player literally shadowing each other. It will mean the variables in the games according to each location at a specific time would be quite close.

Edit!!

Having said that I wish reviewers tested two or three areas in a game.
 
Yeah I suppose.

What’s wrong with running a section of a mp map with the same number of players etc for a 10 min run 3 times, then working out the average.

It will give you a real world scenario in the same way if you were playing the game yourself.
 
Only way to test a multiplayer game is to play it. For example if you want to include bf1 you need multiple plays on a 64 player map. If you played the map enough times with consistently the same number of players you would build up real world results. For reviewers it would be ridiculous as the time needed to play through enough full games to produce usable results would be incredibly time consuming.
 
Statistically mp benchmark is inherently variable and require a vast amount of repeats to get a representative data.

For example, if in the first and second game there’s 20 explosion whereas in the second game there’s 5, do you still take the average? I suppose you can work your way round it by plotting the frame time and find the total low 0.1% and low 1% but it’s statistically challenging to make it reproducible. If there’s a preset mp benchmark (i.e. a prefixed sequencenof event that would happen in a mp player game) then the results would be more representative.

Btw, running around the map three times only represent part of the gameplay, unless the game is only about running around.
 
Another issue is that a lot of reviewers only get samples days before release and don't have time to do loads of repeats that'd be needed for multiplayer games. A lot of the time the hardware is on loan too, they don't even get to keep their samples to do more in-depth testing later on. Then there's the whole issue of content based on new hardware will usually get more views than content based on stuff that came out weeks ago, and you can pretty easily understand why we don't see much multiplayer game testing. I wish that wasn't the case.
 
If you guys look back through this thread 8 pack did say.

I can't remember off top of my head but without delid temps is what stops you going higher.

Edit:

This is what 8 pack wrote.


8 Pack said:
After delid on custom water 1.425-1.45 is fine I think. Aio 240mm around 1.4-1.42.

Our binned are 1.35-1.4 for the given speed. Usually 1.37.
Click to expand...
 
Last edited:
gavinh87 said:
Speaking of reviewers, it seems joker is full of crap.
Click to expand...

When Joker was running benchmarks pitting 1800X vs 7700K @ 720p and Quake 640x480 his benchmarks were received by this very forum with "wow look how a quad core trashes the 8 core. booo AMD. Ryzen is terrible"
now some don't like his benchmarks, now he is full of crap?
 
Panos said:
When Joker was running benchmarks pitting 1800X vs 7700K @ 720p and Quake 640x480 his benchmarks were received by this very forum with "wow look how a quad core trashes the 8 core. booo AMD. Ryzen is terrible"
now some don't like his benchmarks, now he is full of crap?
Click to expand...

That was not said by me. It's not the first time his videos have been disproved.
He had another where the 580 was beating the 1060 by quite a margin. He quickly took that down once he got called out on it.


Besides, were talking about windows performance we don't need another bloody ryzen thread.
 
You must log in or register to reply here.
Back
Top Bottom