DOOM Eternal will run at 1,000 FPS with strong enough hardware

[WYP]

The id Tech 7 engine is ready for next-generation hardware!



Read more about DOOM Eternal supporting 1,000 FPS framerates.

 

[AlienALX]

Wait, it's not April 1st.....

 

[NeverBackDown]

They are claiming 1,000fps but were only able to achieve 400fps in testing. It makes perfect sense. Bold claims really. They couldn't even bother to lower the resolution with min details and try for any higher to prove it works?

 

[WYP]

They are claiming 1,000fps but were only able to achieve 400fps in testing. It makes perfect sense. Bold claims really. They couldn't even bother to lower the resolution with min details and try for any higher to prove it works?

The hard limit in software is 1,000 FPS. Today's hardware can get it to 400 FPS. 1,000 FPS is what future hardware generations will aim for.

DOOM 2016 was limited to a maximum of 250FPS, and now the limit has been stretched higher thanks to engine changes.

 

[AlienALX]

Old hardware can do that too with old games. Like really old games.

 

[demonking]

By the time the hardware catches up to be able to run this at 1000fps at a decent resolution, it will be about as relevant as getting Simon the sorcerer to run at 60fps. Though looking forward to a 1000hz monitor to run this on.....

 

[NeverBackDown]

The hard limit in software is 1,000 FPS. Today's hardware can get it to 400 FPS. 1,000 FPS is what future hardware generations will aim for.

DOOM 2016 was limited to a maximum of 250FPS, and now the limit has been stretched higher thanks to engine changes.

You didn't read what I said. They are claiming something they cannot test or prove. They are just guessing

 

[WYP]

You didn't read what I said. They are claiming something they cannot test or prove. They are just guessing

I think id Software knows its own engine. While they have not "proven" it, they know what their engine will handle on a software level.

The point is that PC hardware and display technologies will need to advance a lot before DOOM Eternal itself becomes a limiting factor.

 

[NeverBackDown]

I think id Software knows its own engine. While they have not "proven" it, they know what their engine will handle on a software level.

The point is that PC hardware and display technologies will need to advance a lot before DOOM Eternal itself becomes a limiting factor.

I didn't say they didn't. It's just a theory they are basing their engine on.

For all they know they could have an algorithmic solution of O(log n) as a worst-case yet in runtime, it could perform far worse than that model would suggest, say O(n^2) as they approach 1000.


I am not saying they are dumb or wrong. I am saying they are making a claim without having physical proof of it and using a Theory-based guess isn't something they should advertise. They should have said, "we have reached 400fps currently with modern top end hardware, but we have designed the engine to work up to 1000fps based on our predictions of what needs to be done to reach that point." Outright claiming something then asking them to prove it, their response would be "we cannot currently do that". Then why claim it? That's like doing a scientific experiment for a drug to help people and when the community asks for proof they say "we don't have any". Yeah, good luck with that!

 

[AlienALX]

It's not dude.

When you code a game you code the engine and thus your rules.

FO4 is coded for 60 FPS. Hacking that breaks accessing computers etc. However if you code the game like a benchmark (no caps balls out) it should be fine.

 

[NeverBackDown]

It's not dude.

When you code a game you code the engine and thus your rules.

FO4 is coded for 60 FPS. Hacking that breaks accessing computers etc. However if you code the game like a benchmark (no caps balls out) it should be fine.

Just because you code it and make up the rules doesn't mean you're exempt from all the limitations of computer theory and algorithms.

I have not even seen a synthetic benchmark that can run 1000 fps. Has anyone done it? I haven't seen that.

 

[AlienALX]

Just because you code it and make up the rules doesn't mean you're exempt from all the limitations of computer theory and algorithms.

I have not even seen a synthetic benchmark that can run 1000 fps. Has anyone done it? I haven't seen that.

Then test it. Download a really old one and try it.

I'm not arguing that it's not totally a gimmick but I'd imagine they know what they're doing.

 

[grec]

I didn't say they didn't. It's just a theory they are basing their engine on.

For all they know they could have an algorithmic solution of O(log n) as a worst-case yet in runtime, it could perform far worse than that model would suggest, say O(n^2) as they approach 1000.


I am not saying they are dumb or wrong. I am saying they are making a claim without having physical proof of it and using a Theory-based guess isn't something they should advertise. They should have said, "we have reached 400fps currently with modern top end hardware, but we have designed the engine to work up to 1000fps based on our predictions of what needs to be done to reach that point." Outright claiming something then asking them to prove it, their response would be "we cannot currently do that". Then why claim it? That's like doing a scientific experiment for a drug to help people and when the community asks for proof they say "we don't have any". Yeah, good luck with that!

Nah, the 200FPS(Later 250fps) cap of idTech6 and 60FPS cap of idTech5 were hard coded limits(IE Limits on the minimum time between updating a frame at 4ms[250fps] or 5ms[200ms]), and they do explicitly state they're referring to increasing the hard coded limit of the engine to 1000FPS(IE Their idTech7 engines hard coded minimum time period between updates is 1ms), none of this has anything to do with algorithmic complexity as their claim explicitly removes the question of power:

On id Tech 6, we maxed it out to 250 frames per second. This game, if you have the hardware right, it could hit 1,000 frames per second. That's the max we have

 

[NeverBackDown]

Then test it. Download a really old one and try it.

I'm not arguing that it's not totally a gimmick but I'd imagine they know what they're doing.

For the 4th time I didn't say they don't know what they are doing. I'm asking for PROOF.

Nah, the 200FPS(Later 250fps) cap of idTech6 and 60FPS cap of idTech5 were hard coded limits(IE Limits on the minimum time between updating a frame at 4ms[250fps] or 5ms[200ms]), and they do explicitly state they're referring to increasing the hard coded limit of the engine to 1000FPS(IE Their idTech7 engines hard coded minimum time period between updates is 1ms), none of this has anything to do with algorithmic complexity as their claim explicitly removes the question of power:

Just because they lower the minimum timeframe doesn't *mean* it's possible to do on the engine. That's just a theory based number. Sure the math checks out, as do many theories, doesn't mean it's feasible. My example of complexity was just an example. Not literal.

Again just as you said too. "their *claim* removes the question of power"
Exactly, a claim. Not a fact.


I don't doubt that lead engineer knows what he's talking about. I just want them to reword it as it's essentially creating fake hype and false advertising until they can show it