GTA V PS4: 1080@30, Core i3/750Ti: 1080@60. How is this possible?

I wonder how cpu threaded GTAVs engine is. I also wonder how things wouldve turned out graphically had they shifted more cpu cycles onto the gpu. Finally, I wonder what the frame rate would be like without the cap at 30.

There's a reason why PS3 first party devs are producing such amazing visuals on PS4. They've been forced to make efficient use of 6 cpu cores for a decade now. With that done, they're free to play with the gpu in impressive ways. I'll continue to maintan that Ken Kutaragi is a genius. Unfortunately it's human nature to take the easy way out of a situation.
 
This is the type of comparison i was talking about i wanted to see. Along with the post above. Don't get why everyone was getting on me for calling those marketing screens being used to compare to PS4's in games screens not a fair or accurate comparison. Looks like i was right.

There are a lot more scenes to compare though.
 
This is the type of comparison i was talking about i wanted to see. Along with the post above. Don't get why everyone was getting on me for calling those marketing screens being used to compare to PS4's in games screens not a fair or accurate comparison. Looks like i was right.

As I said a page back, those aren't running on maximum settings.
Are you saying the promo screens were bullshots?
 
Your post is a salty strawman + bad reading comprehension

the only settings that rockstar revealed for the console versions was what the sliders were set at for the console version (10 21 and 20 out of a 100 maximum for the 3 sliders)
Which is exactly in line with what people were seeing with their own eyes before.

Erm why am I "salty" ? I have been a PC gamer since the bloody Pentium 1 days. The only modern day console I have owned was an Xbox 360 which I used for about a month (to play GTA 4) and then shoved it in the garden shed because it was an awful experience. So I have no "vested" interests in the PS4 being shown to be "all powerful", I honestly don't care about how powerful the PS4 is.

I remember the days of GTA 4 and I remember all the shit talking that was done "oh the consoles must obviously be low settings and all sliders set to zero, my PC can totally do better than that man". I believe the same thing is happening here, people are underestimating the console settings but hey no it must be because the PS4 is so weak ehhh.

I just find it very difficult to accept the premise that an i3 + 750ti can match the visuals of the PS4. Yes I know the cpu in the PS4 is a piece of crap but the GPU is pretty solid (or was for when it was released). That's what makes me think something else is at work here but hey go at me and call me "salty" because you know I betrayed the "PC code" or something.
 
Noticed that as well. Maybe a bad implementation of FXAA is the cause combined with the DOF filter on the PS4. The shots I took and posted here all include FXAA as well but none look as blurry as the PS4 shots.

Does the PS4 version use FXAA? I remember it being quite jaggy, so it's probably the DOF and the jpeg compression.
 
I've shown you screenshot that show the PC high settings compared to the PS4 version. Yet you still refuse to accept that the PS4 version might be running at high settings. If you still feel that way feel to prove me wrong. Why does it bother you so much?

I mentioned AF because in my previous post I used 16xAF. I corrected myself and explained why.
I'm not refusing to believe. I would just love to see a screenshot comparison between high and very high of the specific setting. Let's say for foliage, I would love for a screenshot in the woods comparing high and very high. I've been waiting for an answer for days now since DF still hasn't released their full analysis.
 
I'm not refusing to believe. I would just love to see a screenshot comparison between high and very high of the specific setting. Let's say for foliage, I would love for a screenshot in the woods comparing high and very high. I've been waiting for an answer for days now since DF still hasn't released their full analysis.

candy land might have you covered...

https://youtu.be/v4cbcpmESuA

they also have videos of pc on low, medium and high settings
 
NXGamer released his tech analysis of GTA 5 and made some pretty good comparisons.
https://m.youtube.com/watch?v=QhlLypEJq-c

So in summary from his video:
Tessellation on consoles is comparable to high on PC
Water on consoles is somewhere between high and very high on PC
Vegetation is very high on PS4 but high on XB1
Post FX is somewhere between high and very high. He couldn't tell much difference
Reflections in mirrors are ultra on PS4 but between high and very high elsewhere
Shadows are somewhere between high and very high on consoles with a soft look
Headlights from cars do not cast shadows on most objects on consoles

Also, the GTX 750 Ti runs a lot more like what I expected in this video. Rare minor dips below 30 but mostly hovering around 30-40FPS in car chases.
 
The 8350+7870 does not perform too badly in NXgamer's video but AMD should really step up. All this cpu overhead from their drivers is really killing their framerate on PC. If they sort this out, it seems that the setup could have been closer to a locked 60fps like the i5/970 provides.
 
If there's this much debate on minor differences between the PC version running at the same res on high, I think that says enough. I really was expecting a bigger discrepancy, but that can only be found on a rig I'll never be putting that much money into.
 
NXGamer released his tech analysis of GTA 5 and made some pretty good comparisons.
https://m.youtube.com/watch?v=QhlLypEJq-c

So in summary from his video:
Tessellation on consoles is comparable to high on PC
Water on consoles is somewhere between high and very high on PC
Vegetation is very high on PS4 but high on XB1
Post FX is somewhere between high and very high. He couldn't tell much difference
Reflections in mirrors are ultra on PS4 but between high and very high elsewhere
Shadows are somewhere between high and very high on consoles with a soft look
Headlights from cars do not cast shadows on most objects on consoles

Also, the GTX 750 Ti runs a lot more like what I expected in this video. Rare minor dips below 30 but mostly hovering around 30-40FPS in car chases.
I'll give him the reflections in the mirror being comparative to ultra in the PC version, I didn't check that and assumed it would be similar to the other reflections in the game.

Outside of that I don't agree with a lot of his "seems to be between high and very high" guesses. If you can't tell the difference or don't find the difference big enough that doesn't automatically makes it the higher option.

This is the first time I've seen one of this vids, does he normally benchmark games with vsync enabled?
 
I wouldn't touch the G3528, but the i3 did really well there.

Yeah the Alienware was a great deal at $400. I just wish you could upgrade the GPU.

An i3/i5 paired with a 750Ti only needs a 250-280 watt power supply, so you could almost definitely add to an existing home PC. Granted it would need a PCI express slot.



Yeah, You basically have an i7 Alienware Alpha. The i3 base model has been very competitive with the PS4.

Actually, I have a G3528 coupled with a 750ti and GTA V runs fine at 1080P here.
 
I'll give him the reflections in the mirror being comparative to ultra in the PC version, I didn't check that and assumed it would be similar to the other reflections in the game.

Outside of that I don't agree with a lot of his "seems to be between high and very high" guesses. If you can't tell the difference or don't find the difference big enough that doesn't automatically makes it the higher option.

This is the first time I've seen one of this vids, does he normally benchmark games with vsync enabled?

Like ac unity...What's not on ps4 is not enough visible,worth the performance cost or useless....
Man if you're not biased from start you can't take this guy seriously.
 
I'll give him the reflections in the mirror being comparative to ultra in the PC version, I didn't check that and assumed it would be similar to the other reflections in the game.

Outside of that I don't agree with a lot of his "seems to be between high and very high" guesses. If you can't tell the difference or don't find the difference big enough that doesn't automatically makes it the higher option.

This is the first time I've seen one of this vids, does he normally benchmark games with vsync enabled?
I'm not sure, but the point of this video is not to show how stable the lock is, but to show how well each PC runs the game. And BTW, no one said it would automatically mean very high by default if the difference is hard to catch. Who said that? High to very high could mean higher in some areas and lower in others or somewhere in between the two in all areas. And to be fair, you never said the reflections in mirrors were comparable to ultra and vegetation was comparable to very high on PS4 even when I asked again.

Like ac unity...What's not on ps4 is not enough visible,worth the performance cost or useless....
Man if you're not biased from start you can't take this guy seriously.
Just as how DF is biased towards XBox and most people have no problems trusting them?
(No offense, John, you're the only one who is neutral in your tech analysis tbh)
 
Is texture quality actually a performance variable or mainly down to how much ram you have available? I.e given identical setups, both with enough video memory to host the textures, would a very high texture take more effort to draw? If not then the choice of textures on PS4 would be mainly down to available memory rather than performance?

So even if you found out the PC equivalent detail settings and set the GTX750 to that, it only has 2GB ram so that might make it perform worse than PS4 because of the lack of ram. Or would it be able to transfer textures across from main memory fast enough to compensate?
 
Is texture quality actually a performance variable or mainly down to how much ram you have available? I.e given identical setups, both with enough video memory to host the textures, would a very high texture take more effort to draw? If not then the choice of textures on PS4 would be mainly down to available memory rather than performance?

So even if you found out the PC equivalent detail settings and set the GTX750 to that, it only has 2GB ram so that might make it perform worse than PS4 because of the lack of ram. Or would it be able to transfer textures across from main memory fast enough to compensate?

Texture quality has barely any performance hit.
http://www.gamersnexus.net/game-bench/1909-gta-v-graphics-optimization-guide

gta-v-texture-fps.jpg


However, if you don't have enough VRAM then stuttering can occur. If you have enough then it does not matter.

In GTA 5 2gb is enough for max texture quality.
 
NXGamer released his tech analysis of GTA 5 and made some pretty good comparisons.
https://m.youtube.com/watch?v=QhlLypEJq-c

So in summary from his video:
Tessellation on consoles is comparable to high on PC
Water on consoles is somewhere between high and very high on PC
Vegetation is very high on PS4 but high on XB1
Post FX is somewhere between high and very high. He couldn't tell much difference
Reflections in mirrors are ultra on PS4 but between high and very high elsewhere
Shadows are somewhere between high and very high on consoles with a soft look
Headlights from cars do not cast shadows on most objects on consoles

Also, the GTX 750 Ti runs a lot more like what I expected in this video. Rare minor dips below 30 but mostly hovering around 30-40FPS in car chases.
That seems be because of his CPU choice in the video (horrible AMD cpu). Also, PC version uses a different style DOF (Bokeh).
 
That seems be because of his CPU choice in the video (horrible AMD cpu). Also, PC version uses a different style DOF (Bokeh).
Yeah, I'm aware of that, but NXGamer's video proves that the PS4 version is not entirely on high settings. I tried asking others in this thread but never got a good answer. And it would be interesting to see the differences between that AMD CPU and an i3 CPU. Hopefully DF's full analysis uses the latter, but in gameplay areas, unlike their first analysis.
 
That is interesting. But then again... the texture quality in this game is just OK IMO.
Its not just the quality of each individual texture than matters, i think in this game it's mostly the sheer quantity of different textures that are shown at any given scene.
 
Really? The PS4 should easily be able to have textures at very high, then. But from what I heard in this thread, it seems to be a mix of high and very high.
Theoretically it should be at very high. 2gb of VRAM is enough if you lower the distance scaling slider to half. It eats VRAM like it's nothing.

That is interesting. But then again... the texture quality in this game is just OK IMO.
Well for a game of that scale I'm really impressed by the texturework.
From what I've seen very high textures are not more detailed than on Xbox One and PS4 which is to be expected since they don't require too much video memory.
With maxed distance scaling and very high textures VRAM is at 2390mb.

And it would be interesting to see the differences between that AMD CPU and an i3 CPU. Hopefully DF's full analysis uses the latter, but in gameplay areas, unlike their first analysis.
Is that enough for you :
gtav_vhigh_cpu.png

8ORihEm.jpg
 
Theoretically it should be at very high. 2gb of VRAM is enough if you lower the distance scaling slider to half. It eats VRAM like it's nothing.


Well for a game of that scale I'm really impressed by the texturework.
From what I've seen very high textures are not more detailed than on Xbox One and PS4 which is to be expected since they don't require too much video memory.
With maxed distance scaling and very high textures VRAM is at 2390mb.
Yeah, I'm thinking the textures are on very high as well. Even the very high setting has some ugly textures here and there. Perhaps this is the reason which led some to believe it's below very high.
 
Yeah, I'm thinking the textures are on veey high as well. Even the very high setting has some ugly textures here and there. Perhaps this is the reason which led some to believe it's below very high.

My expectations is that consoles use a mix of high and very high settings.
I still expect a 750ti paired with a decent I5 to match consoles though.

As far as I'm concerned GTA 5 on PC is the absolute best version of the game I've played (PS3-PS4-PC). Flawless would be putting it lightly.
Hardware permitting it definitely has effects not present on consoles : PCSS, better draw distance, better LOD and probably other things.
 
My expectations is that consoles use a mix of high and very high settings.
I still expect a 750ti paired with a decent I5 to match consoles though.

As far as I'm concerned GTA 5 on PC is the absolute best version of the game I've played (PS3-PS4-PC). Flawless would be putting it lightly.
Hardware permitting it definitely has effects not present on consoles : PCSS, better draw distance, better LOD and probably other things.
Yep, no doubting that. But the biggest improvement would be 4K resolution IMO. Still, the PS4 version holds up reasonably well. And I don't know how much you trust NXGamer, but I made a summary of his tech analysis earlier, which shows the PS4 version is a mix of high and very high.
 
It looks great but it's not the posterboy for 4K in my opinion. Ryse or even Unity are.
I see your point. In this case, I agree. Just out of curiousity, is it possible for grass to be completely absent on PC just like the last gen versions?
 
I see your point. In this case, I agree. Just out of curiousity, is it possible for grass to be completely absent on PC just like the last gen versions?

I guess so if you set it all the way down to normal. I have not checked and considering you need to restart the game for some effect to engage, I don't feel like doing so.

This makes no sense in this situation.
Ignore him, just trolling for attention.
 
Top Bottom