Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

AppGameKit Studio Chat / Speed test Vulkan vs Basic vs Classic

Author
Message
Pfaber1
6
Years of Service
User Offline
Joined: 7th Jan 2018
Location: England
Posted: 18th May 2021 08:52
Moving a Cube from left to right in a loop with fps set to 600 these are my results .

Studio
Vulkan average 290 fps
Basic average 190 fps

Classic average 190 fps

Here's my code If you can see any flaws in my method please let me know.

Raven
19
Years of Service
User Offline
Joined: 23rd Mar 2005
Location: Hertfordshire, England
Posted: 18th May 2021 13:56
Both BASIC and ADVANCED run at 600 FPS.
Changing the Sync Rate to 0

(Studio)
BASIC ~ 2700 FPS
ADVANCED ~ 3600 FPS (1900 FPS when Overlays are On-Screen)

(Classic)
BASIC - 2400 FPS

Vulkan is only showcasing 33% Better Performance in Studio, and the difference between Classic and Studio is 12.5%
Neither are close to the 2x Performance you were claiming.... now aside from that., well the test is far too limited in scope.
Pfaber1
6
Years of Service
User Offline
Joined: 7th Jan 2018
Location: England
Posted: 18th May 2021 15:11 Edited at: 18th May 2021 15:31
Well here it is I set the sync rate to 0 and in Basic I got about 200 and in Vulkan I get 1100 to 1200. I don't know why this is but there it is , I tested several times .
My laptop is HP 15s fq1020na 8gigs of ram ,intel i3 1.2 to 3.4 gig processor . Forgot to mention it runs much faster when it's connected to a power source.
That's six times faster between Basic and Advanced.
Loktofeit
AGK Developer
15
Years of Service
User Offline
Joined: 21st Jan 2009
Location: Sarasota, FL
Posted: 18th May 2021 15:36 Edited at: 18th May 2021 15:38
Changed to SetSyncRate(0,1) to remove caps.

(Classic)
BASIC ~4200 FPS

(Studio)
BASIC ~4200 FPS
ADVANCED ~3700 FPS

Test System Used
CPU: Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz (8 CPUs), ~3.6GHz
RAM: 32768MB RAM
GPU: NVIDIA GeForce GTX 1080 Ti
LynxJSA's web games/quizzes - LynxJSA's Android apps
AGK Resource Directory
"Stick to a single main loop (DO...LOOP) and loop through it every frame.
Do everything inside functions.
Use finite state machines to control your game.
Use lots and lots of source files.
Use virtual resolution instead of the default percentage system." - Digital Awakening
jd_zoo
5
Years of Service
User Offline
Joined: 12th May 2018
Location: Nova Scotia
Posted: 18th May 2021 21:02
Fluctuates rapidly but after stabilizing a bit these are my rough numbers (although a bit dizzy from watching that...)

(Classic)
BASIC ~2700 FPS

(Studio)
BASIC ~2700 FPS
ADVANCED ~3900 FPS

AMD Ryzen 5 3600 6-Core 3.59GHz, 16GB RAM
GeForce RTX 2060
SFSW
21
Years of Service
User Offline
Joined: 9th Oct 2002
Location:
Posted: 18th May 2021 22:55
All appears to be within what I'd expect for results (Intel vs AMD, age of GPU's, example code, etc), but I wouldn't put much weight in a simple primitive example where you may find GL can do better on some configurations. The benefits of Vulkan generally surface most with systems better optimized for it (generally newer) and seem to compound further the more shader/graphic effects you throw at it.
Pfaber1
6
Years of Service
User Offline
Joined: 7th Jan 2018
Location: England
Posted: 19th May 2021 07:22
I just tried it using my latest project and the lowest results I got were Basic 60 and Advanced 110 . This is using battery power on my laptop and it's not a gaming rig just a middle of the road computer .
Loktofeit
AGK Developer
15
Years of Service
User Offline
Joined: 21st Jan 2009
Location: Sarasota, FL
Posted: 19th May 2021 11:54
When you post these test results, could you also post the test that was used? It would both add some context to the results and also allow others to run the same tests for a wider sample base.
LynxJSA's web games/quizzes - LynxJSA's Android apps
AGK Resource Directory
"Stick to a single main loop (DO...LOOP) and loop through it every frame.
Do everything inside functions.
Use finite state machines to control your game.
Use lots and lots of source files.
Use virtual resolution instead of the default percentage system." - Digital Awakening
Raven
19
Years of Service
User Offline
Joined: 23rd Mar 2005
Location: Hertfordshire, England
Posted: 19th May 2021 16:06
I think a Benchmarking tool might be a good idea to put together., as I find the results (limited as they are) interesting.
Now as a note... Pfaber., I'm sorry but your Intel Iris HD Graphics are "Below Minimum" in terms of Performance... as _most_ Mobile Devices will actually be capable of better performance metrics.

But beyond that, it would be interesting to put together a benchmark test and get a bigger sample size.
As it stands a:
Ryzen 5 1600 / RX 5700 XT
Ryzen 5 3600 / RTX 2060
Intel iCore 7 4790K / GTX 1080 Ti

Those are certainly setups towards the higher end of Mainstream but with some interesting results.
Think I'll have a tinker this evening, and put something together.
Pfaber1
6
Years of Service
User Offline
Joined: 7th Jan 2018
Location: England
Posted: 19th May 2021 17:10
Yes that would be very interesting . I know my setup is a bit feeble at best but it's probably average for what most people have . Look forward to trying your tests presuming I'll be able to run them . sorry Loktofeit I can't post my half finished project on here but I will bare what you said in mind. I've had this laptop about 4 months and bought it new . Seems to do much better with Vulkan 1.2 than it does with openGL 4.5
Loktofeit
AGK Developer
15
Years of Service
User Offline
Joined: 21st Jan 2009
Location: Sarasota, FL
Posted: 20th May 2021 00:30
"it's probably average for what most people have" - related to that, I was recently looking at Steam stats. Steam does regular surveys, so if we accept that the average general gamer may have a system something like the average Steam gamer, here are Steam's top results for each category for April 2021:

OS Version - Windows 10 64 bit - 92.83%
System RAM - 16 GB - 45.52%
Intel CPU Speeds - 3.3 Ghz to 3.69 Ghz - 17.52%
Physical CPUs - 4 cpus - 40.77%
Video Card Description - NVIDIA GeForce GTX 1060 - 8.97%
VRAM - 8191MB - 24.27%
Primary Display Resolution - 1920 x 1080 - 67.60%
Microphones - Yes - 100.00%
Free Hard Drive Space - 100 GB to 249 GB24.31%+0.25%
Total Hard Drive Space - Above 1 TB53.04%+0.37%
VR Headsets - Oculus Quest 227.79%+3.53%

Figured that might be useful data for AppGameKit devs making game performance decisions and setting minspecs.

Source: https://store.steampowered.com/hwsurvey/Steam-Hardware-Software-Survey-Welcome-to-Steam
LynxJSA's web games/quizzes - LynxJSA's Android apps
AGK Resource Directory
"Stick to a single main loop (DO...LOOP) and loop through it every frame.
Do everything inside functions.
Use finite state machines to control your game.
Use lots and lots of source files.
Use virtual resolution instead of the default percentage system." - Digital Awakening

Login to post a reply

Server time is: 2024-03-29 12:10:12
Your offset time is: 2024-03-29 12:10:12