Should we drop 5760x1080 in VGA reviews?

[Breaz]

and it's decided .. dropping 5760x1080 and replacing 2560x1600 with 2560x1440.

-> 1600x900, 1920x1080, 2560x1440, 3840x2160

list of games:

great this is fantastic

 

[patrico]

looks good to me

 

[chuck216]

Since I'm poor and couldn't afford anything better than a 1680 x 1050 monitor I find any resolution over that to be useless.

 

[HammerON]

and it's decided .. dropping 5760x1080 and replacing 2560x1600 with 2560x1440.

-> 1600x900, 1920x1080, 2560x1440, 3840x2160

list of games:

I am sad to see 2560x1600 be replaced by 2560x1440, however I understand that there are many more people using the 1440 res.

 

[HumanSmoke]

Since I'm poor and couldn't afford anything better than a 1680 x 1050 monitor I find any resolution over that to be useless.

Well, if W1zzard was providing these performance reviews solely for your benefit or it was found that that a sizeable number of readers also used that resolution, your comment would carry some relevance.

Or was your post an attempt at begging for donations for a higher res screen?

 

[chuck216]

Well, if W1zzard was providing these performance reviews solely for your benefit or it was found that that a sizeable number of readers also used that resolution, your comment would carry some relevance.

Or was your post an attempt at begging for donations for a higher res screen?

No, more a failed attempt at humor.

 

[Frag_Maniac]

If you want to add a widescreen resolution you should consider 3440x1440 from the new 34" 21:9 monitors. They are quite "popular". Which means they are very popular in forums but few people have them.

As I said above, I think the 21:9 displays will eventually be preferred over and replace a lot of triple display setups, but I don't think we're quite at the point to include them in mixed benches.

Since I'm poor and couldn't afford anything better than a 1680 x 1050 monitor I find any resolution over that to be useless.

I look at your spec list and I'm thinking your not being able to afford one is a bit of an exaggeration. The key word seems to be "couldn't", not can't. How long ago was the 1050p display bought?

I'm willing to bet since purchasing it, the 1080p displays have been halved in priced. A simple solution for now is just to combine the FPS for 1080p and 900p and divide them in two, which is almost exactly the number of pixels you're using.

These are the kind of inconveniences you have to deal with if you don't stay up on current hardware trends and want/expect exact benches results.

 

[newconroer]

No plans for 1440p, my monitor is 1600p.

W1zzard, you should get a hold of a 1440p Korean Catleap monitors and crank it up to 120hz. That way you can do reviews with 1440p as well as 120hz, which could effectively replace the 5870 aspect of the benchmarks.
It's a shame after all the publicity TPU brings manufacturers, that someone hasn't dropped you a 1440p IPS panel yet.

 

[newtekie1]

W1zzard, you should get a hold of a 1440p Korean Catleap monitors and crank it up to 120hz. That way you can do reviews with 1440p as well as 120hz, which could effectively replace the 5870 aspect of the benchmarks.
It's a shame after all the publicity TPU brings manufacturers, that someone hasn't dropped you a 1440p IPS panel yet.

He already addressed this. He doesn't use vsync when benching, so the refresh rate of the monitor is irrelevent, and his monitor is already capable of 2560x1440.

 

[newconroer]

He already addressed this. He doesn't use vsync when benching, so the refresh rate of the monitor is irrelevent, and his monitor is already capable of 2560x1440.

Of course it's relevant. A 3d program that doesn't have a capped/consistent frame timings/latency is going to produce wildly varying results and ultimately tell us nothing.

I never noticed the number to the right of the decimal on his FPS results. Is he just showing maximum frames?


Maybe the monitor is capable, but that's not a native resolution - and that makes a difference,

 

[cadaveca]

Is he just showing maximum frames?

TPU policy is to use averages. In everything, unless explicitly stated with the result (ie, Max power consumed).

 

[chuck216]

As I said above, I think the 21:9 displays will eventually be preferred over and replace a lot of triple display setups, but I don't think we're quite at the point to include them in mixed benches.I look at your spec list and I'm thinking your not being able to afford one is a bit of an exaggeration. The key word seems to be "couldn't", not can't. How long ago was the 1050p display bought?

I'm willing to bet since purchasing it, the 1080p displays have been halved in priced. A simple solution for now is just to combine the FPS for 1080p and 900p and divide them in two, which is almost exactly the number of pixels you're using.

These are the kind of inconveniences you have to deal with if you don't stay up on current hardware trends and want/expect exact benches results.

Believe it or not, the entire system with the exception of the monitor and case was a gift to replace a 5 year old system that had an unfortunate encounter with a lightning strike,,

I have a generous Significant Other that does freelance work for a computer shop.

I didn't pay a cent for my system

 

[Sinzia]

As someone who games at 5760x1080 in a few games, I'll just look to the 4k numbers and approximate it from there.

 

[LAN_deRf_HA]

W1zzard, you should get a hold of a 1440p Korean Catleap monitors and crank it up to 120hz. That way you can do reviews with 1440p as well as 120hz, which could effectively replace the 5870 aspect of the benchmarks.
It's a shame after all the publicity TPU brings manufacturers, that someone hasn't dropped you a 1440p IPS panel yet.

And make blood pour from his eyes? Overclocking those 1440p displays makes them drop color information, the faster the worse it is, and no a color profile doesn't fix it. It's a deep, eye rotting effect. If you really want to go faster on 1440p with IPS you have to wait for this http://techreport.com/news/27019/au-optronics-panel-to-combine-ips-tech-144hz-refresh

 

[GhostRyder]

We don't need to have all the resolutions out there to get an understanding of what we can expect so I am all for dropping it even though I still from time to time do it.

As long as there are the basic resolutions available as a single screen then we should all be good. 1600x900, 1920x1080/1200, 2560x1440/1600, and 3840x2160 we can all get a good estimate of our numbers. Heck even below 1080p could be dropped almost at this point since that has become the standard at this point if needed at least I think though many laptops and such still come at that resolution.

 

[Frag_Maniac]

I have a generous Significant Other that does freelance work for a computer shop.

Then that person should be able to get you an exceptional deal on a 1080p monitor. Surely you must have at least SOME money to spend?

I really think you'e making this sound impossible.

 

[chuck216]

Then that person should be able to get you an exceptional deal on a 1080p monitor. Surely you must have at least SOME money to spend?

I really think you'e making this sound impossible.

I never said I wanted a higher resolution monitor though.

I'm quite happy with the one I have until it goes bad.

I'm guessing you didn't see my post about me intending that statement to be a joke.

 

[GhostRyder]

I never said I wanted a higher resolution monitor though.

I'm quite happy with the one I have until it goes bad.

I'm guessing you didn't see my post about me intending that statement to be a joke.

Don't worry, most of us saw that being an obvious joke

Now then @W1zzard, maybe you should replace the 5760x1080 with 7680x1440 (sarcasm intended).

 

[newtekie1]

Of course it's relevant. A 3d program that doesn't have a capped/consistent frame timings/latency is going to produce wildly varying results and ultimately tell us nothing.

Incorrect. If the average is over 120FPS, then it is pretty safe to assume the game will stay pretty steady at 120FPS on a 120Hz panel. If the average is over 144FPS, then it will run at 144FPS most of the time on a 144Hz panel. There might be some dips below that, but even with a native 120/144Hz panel, that isn't going to show up in the review anyway the same way the lows aren't in the reviews now.

Maybe the monitor is capable, but that's not a native resolution - and that makes a difference,

When performance numbers are concerned, it makes absolutely no difference. You are going to get the exact same performance numbers running 2560x1440 on a 2560x1440 monitor and a 2560x1600 monitor.

 

[theJesus]

It would be an easier work load on you but in due honesty cards like the *95/x2/titan/r9*90/*80ti should get a flash by at those res for games that do support it, it would show maximum capability.

I think in the normal reviews 5760x1080 should be dropped, but I think it should be in the SLI/Crossfire reviews. And not really as a comparison to the single cards, so you don't have to bench every card with that resolution, instead maybe just a single page with the performance numbers for just the SLI/Crossfire in all the benchmarks.

I say this because I do think there are people out there doing SLI/Crossfire that are looking at 5760x1080. And, IMO, you should only be running surround with multiple GPUs anyway. So you should present the information, but you don't have to waste time benching every single card.

It should be noted that 4k is more pixels than 5760x1080, so there really is no point to 5760x1080 being in the reviews as far as performance is concerned.

5760x1080=6220800
3840x2160=8294400

If the numbers are acceptable at 3840x2160, they'll be even more acceptable at 5760x1080.

 

[Frag_Maniac]

I never said I wanted a higher resolution monitor though.

I'm quite happy with the one I have until it goes bad.

I'm guessing you didn't see my post about me intending that statement to be a joke.

Actually I saw that, but for some reason thought you were referring to the person's comment you were responding to as a bad attempt at a joke.

It's hard enough to interpret people's meaning as is on forums, and your having to explain your meaning makes that all the more obvious. Emoticons aren't hard to use you know.

 

[newtekie1]

It should be noted that 4k is more pixels than 5760x1080, so there really is no point to 5760x1080 being in the reviews as far as performance is concerned.

5760x1080=6220800
3840x2160=8294400

If the numbers are acceptable at 3840x2160, they'll be even more acceptable at 5760x1080.

Sure, but something might not be acceptable at 3840x2160, but is entirely acceptable at 5760x1080. Not to mention even though it has less pixels, sometime triple monitor setups just behave strangely and have really horrible performance because, being less popular, sometime game developers just don't bother to optimize for that resolution. If performance was all about pixel count, we could just have one benchmark at 1024x576 and use a calculator to figure out the rest...but we all know that doesn't work.

 

[cadaveca]

Sure, but something might not be acceptable at 3840x2160, but is entirely acceptable at 5760x1080. Not to mention even though it has less pixels, sometime triple monitor setups just behave strangely and have really horrible performance because, being less popular, sometime game developers just don't bother to optimize for that resolution. If performance was all about pixel count, we could just have one benchmark at 1024x576 and use a calculator to figure out the rest...but we all know that doesn't work.

Sounds like game problems, not VGA problems. Either way, it wouldn't be indicative of VGA performance.

 

[Mindweaver]

I agree drop it as well. I have an Nvidia Surround setup, but you really need a dual card setup to really enjoy it. My GTX 680 does a pretty good job on low to medium settings. Plus, we are all going to be playing on VR headset's in the next year or two.

 

[Mussels]

for the 'insane' benchmark, run 3x 4K monitors for the SLI/crossfire reviews

just hook the one monitor up with three cables and make it think theres three of em.