CPU Benchmark in realistic condition (multiple programs running)

[coercitiv]

I feel that many here on the forum as well as Anandtech's reviewers are perhaps of the older demographics that might be out of touch with what younger (15-30 yo) gamers do with their computers.

Go to twitch.tv and see what those streamers do. As you get better and better at online competitive games, the queue time often gets longer. What some people do is play another game while waiting. So they often have stuff like Fallout 4, Overwatch, Heathstone being paused in the background while they're playing Dota or LoL. That, on top of streaming, browser open to interact with viewers, skype-ing with teammates in the games, recording the gameplay for review and upload to youtube, the list goes on.

I feel that most young gamers who multitask heavily know better than to try this kind of behavior on systems with dual core CPUs. Otherwise it's not only senior forum members on this forum who are out of touch with modern gaming needs.

Meanwhile, for every successful streamer / youtuber who does all the things you describe, there are hundreds or thousands of subscribers who watch their PCs like television sets and make extensive use of chat windows: these are the people who should consider Pentiums and i3s for their computing needs.

 

[turn_pike]

feel that most young gamers who multitask heavily know better than to try this kind of behavior on systems with dual core CPUs. Otherwise it's not only senior forum members on this forum who are out of touch with modern gaming needs.

The question raised and the response is asymmetric to each other.
Young gamers as well as old one rely on benchmarks to gauge the cpu's performance, they know what they want to do but only have vague notion how well a new cpu can cope with the task.
Its not "modern gaming needs" which is actually pretty low since many would just play MOBAs, CS:GO or Hearthstone but what they do WHILE gaming which might increase the cpu load significantly.

This is worse for aforementioned younger, more social-media-happy gamers since benchmarks typically are done in isolation. While the intentions are good and reasonable (controlled environment, reproducible, isolate variables, etc), it is also often not representative to the real world condition. So often there's no way for them to know beforehand except some vague ballpark estimation. What happens is normally they just try it out themselves or ask people with similar situation (thats why you often see people upload short youtube gameplay, giving their system specs and graphic quality settings as well as mentioned whether they're streaming while doing it).

Meanwhile, for every successful streamer / youtuber who does all the things you describe

While not mainstream, this is actually more common than you'd think.
People with competitive drive often record their gameplay for review and also to stream or upload it to YT to gather helpful advice on how to improve. If you are in a clan, you can watch each other stream to learn how your teammates think and improve your teamwork. You dont need or necessarily want 1000 viewers, just that 3 guys in your clan who is interested to learn from or to teach you.

Even when you dont stream you still multitask, play other games during queue, train your mechanics in Osu or Aimbooster, listen to pandora in background, skype with your buddies, etc etc. Maybe it doesnt matter because modern Pentium or i3 duo-core is so fast but at this point I just dont know

Anyways. I dont think including "while streaming" is necessarily the way benchmarks should go but it does illustrate nicely how different the needs of modern, young gamers are to the way benchmarks are conducted which, at least judging from the response in this thread, fits better for older generation of gamers.
.

 

[sm625]

Since 2003, 98% of my game time has been in windowed mode. And almost always there is a bunch of other stuff running.

I tried to simulate a real world heavy multitasking type scenario for my reviews of intel integrated graphics. The results were rather disastrous. I'd like to repeat the same test with skylake IGP, but I dont hold high hopes for it.

 

[Rakehellion]

So I was reading yet another vociferous debate on the merit of intel vs amd cpus here on AT forum. Amongst the mudslinging there were a couple of interesting points being raised:

- Modern games requiring more and more thread; 4 years old i7 2600 can sometimes outcompete modern i5
- Most benchmarks are done in sterile fashion to prevent confouding variable so almost no other programs are running which can be unrealistic

Now I dunno about most gamers but I usually have multiple programs running while gaming. Firefox and Chrome with double digit tabs each or Torrent, IRC, IMs, Skype, SSH, Netlimiter, Virtualbox and so on. Most of the time they wont take much cpu power but they do add up. 20% or so cpu used is rather common.

Moreover streaming or encoding your play to upload it later to youtube is something that is becoming more and more common. Sometimes I even record myself playing in order to review it later and learn from my mistakes (playing mobas or RTS).

Just having a "dirty benchmark" in general would be more useful.
I hardly ever stop my other programs before gaming, at most I would set the cpu affinity when encoding / compiling. I would think many enthusiast here in AT would feel the same. Most people at least dont quit their browsers while gaming.

With that in mind, is there a website somewhere that test for things like this ?
I'd love to know if i3s and Pentiums are still be the cpu of choice if you want to game and stream for cheap ?
How would the benchmark results look if multiple programs are running in background and you're also streaming your play ?

A benchmark is supposed to test your hardware, nothing more. If you have tabs open/encoding video/50 different viruses eating away at resources, that's your fault.

 

[Rakehellion]

That is missing the point.
What I'd like to know is if they still are good while streaming and recording your gameplay. There's also 30 tabs browsers and multiple programs running in the background (unlike benchmark scenario which is usually clean windows install).

If you want to know, then do a test.

 

[Rakehellion]

No, but I often instal, unzip stuff, update etc while playing games.

If you care about game performance, then stop doing those things. If performance is fine, then you don't need a benchmark.

 

[superunknown98]

I think what OP is trying to ask in more practical terms: Might the extra threads available in an AMD APU provide higher frame rates or more responsive secondary tasks than a dual core Intel chip while running background tasks.

I think it's a valid question.

 

[frozentundra123456]

If you care about game performance, then stop doing those things. If performance is fine, then you don't need a benchmark.

Bingo.

 

[moonbogg]

I like this idea. I say we all benchmark the same game while encoding a video and see who comes out on top. Its like, your Skylake gets high FPS, but what happens if you make it do THIS at the same time! Baha now it sucks.

 

[sandorski]

I always have many apps(although usually low cpu/gpu utilization ones) running, multiple Tabs open(some open for weeks that I haven't used for awhile, but may need again), and other things running. Occasionally I'll have 2 games running, mostly because I forgot one was running when I started a new one.

However, I don't really see the value of a Dirty Benchmark. On the occasion that a Game seems to be not playing as well as usual(usually because another game is running), then I'll check to see what I have running and close certain things. I don't really need a Dirty Benchmark to tell me what to expect. Just give me a Clean one, I'll adjust my situation as needed.

That said, it might make for an interesting Article for some website to publish. Contrast it with the Good Old Days and use it as a investigation into how far we have come. 15+ years ago I remember having to do all sorts of things to eek out extra Frames for many games, close certain Apps, Free Memory, and the like. These days, by contrast, I just start up whatever I want without a second thought and 95+% of the time everything is perfectly fine, even when I discover afterwards that I had another game running the whole time. Which was completely impossible 15+ years ago.