*** Cyberpunk 2077 ***

Set the Texture Filtering -LOD Bias to -3 in Nvidia Profile Inspector as that sharpens the image, and I don't really like Reshade. I use the same (plus other settings) for ETS2 and it helps a ton, but not all the same settings go across all games sadly.

https://i.imgur.com/5eGsGOp.jpg

When the modding tools out I never really took much interest in them, but seeing what we have now I might get back into it. Now that we have a repacker available it might just be something I can tackle.
 
Pete @r... said:
I know pc gaming isn't a closed system like consoles, lots of permutations, but games in years gone by (I'm old) were never released this bad
Click to expand...
... Skyrim?

Not to bash your point though; I remember when games shipped on floppy or (ZOMG) CD-ROM and they were properly bloody finished when they did. Just Cyberpunk is hardly the worse AAA offender in the last decade IMHO.

Anyway: gonna try 1.2 Woo-hoo!
 
I am just glad I can sleep normally

I would not expect any big changes to pedestrian AI anytime soon. If they even will want to overhaul it, it would be a longterm (year plus) project. AI for open world in modern city is insanely hard thing, and hardware demanding. And the game barely runs on lastgen as it is.

Kingdom Come has the kind of NPC AI that people expected from Cyberpunk. But Kingdom Come takes place in a countryside with orders of magnitude less NPCs...and also barely runs on lastgen.
 
philo-sofa said:
... Skyrim?

Not to bash your point though; I remember when games shipped on floppy or (ZOMG) CD-ROM and they were properly bloody finished when they did. Just Cyberpunk is hardly the worse AAA offender in the last decade IMHO.

Anyway: gonna try 1.2 Woo-hoo!
Click to expand...

Definitely not saying it's the worse, just that it seems to happen a lot now a days. Games were finished back then, now it seems they have a dead line and whether it's finished or not, release it. Shocking!
 
Pete @r... said:
Definitely not saying it's the worse, just that it seems to happen a lot now a days. Games were finished back then, now it seems they have a dead line and whether it's finished or not, release it. Shocking!
Click to expand...

Take this with a pinch of salt but I worked a summer at EA and then Lionhead (gives you an idea how long ago!) in their QA department and what they said they do at the time (bearing in mind at this point physical copies were still pretty dominant) is as follows. You take the "size" of the game (something like the total number of lines of code; it's flawed but you need to estimate somehow) and they have some ratio based on previous releases or some other stats which tells them approximately how many bugs they might reasonably expect the game to have. Then you start your QA process and keep going until you have found a certain % of that estimate of the bugs, you assume that ought to contain most/all of the most serious bugs and release the result. We weren't told this but I expect when there are pressures to release and whatnot perhaps they lower that percentage to try and get it out of the door sooner, and of course now that games can be patched they are probably far more lenient in what they will call "finished"
 
It happened back in the day as well - but a lot more games these days are produced with at least at some level a lack of any passion or pride in the final product - whether that is developer or publisher or a mixture of both.
 
uncle_rufus said:
Take this with a pinch of salt but I worked a summer at EA and then Lionhead (gives you an idea how long ago!) in their QA department and what they said they do at the time (bearing in mind at this point physical copies were still pretty dominant) is as follows. You take the "size" of the game (something like the total number of lines of code; it's flawed but you need to estimate somehow) and they have some ratio based on previous releases or some other stats which tells them approximately how many bugs they might reasonably expect the game to have. Then you start your QA process and keep going until you have found a certain % of that estimate of the bugs, you assume that ought to contain most/all of the most serious bugs and release the result. We weren't told this but I expect when there are pressures to release and whatnot perhaps they lower that percentage to try and get it out of the door sooner, and of course now that games can be patched they are probably far more lenient in what they will call "finished"
Click to expand...

I work in prediction and their methodology seems all wrong. You’d want to chart the rate at which you find bugs, then chart bugs found per week, forecast it forward to total number of bugs then stop once you reach an acceptable number of bugs.

The act of forecasting would show the true number of bugs in a game, and illustrate how good the QA testers were.

If you do it their way you could have a squillion bugs and stop.
 
philo-sofa said:
I work in prediction and their methodology seems all wrong. You’d want to chart the rate at which you find bugs, then chart bugs found per week, forecast it forward to total number of bugs then stop once you reach an acceptable number of bugs.

The act of forecasting would show the true number of bugs in a game, and illustrate how good the QA testers were.

If you do it their way you could have a squillion bugs and stop.
Click to expand...

I agree, it wasn't a good system at all :D
 
philo-sofa said:
I work in prediction and their methodology seems all wrong. You’d want to chart the rate at which you find bugs, then chart bugs found per week, forecast it forward to total number of bugs then stop once you reach an acceptable number of bugs.

The act of forecasting would show the true number of bugs in a game, and illustrate how good the QA testers were.

If you do it their way you could have a squillion bugs and stop.
Click to expand...

Developers/publishers use all kinds of approaches, mostly BS, for this kind of thing - but game code just doesn't work like that. For instance the source of 2 games could be equal size but say one of them has dozens of monsters which all use variants of the same underlying code which is already robust and minimal chance for further bugs while the other has a complex feature such as perk mechanics that touches almost every part of the game using the same number of lines of code with the potential for 100s of bugs.
 
Quick little play before work, vehicles handle a lot better but feel slow, I don't know if that's my perception as I've not played for a couple months but flat out in the caliburn feels positively pedestrian now.
 
Sp00n said:
Quick little play before work, vehicles handle a lot better but feel slow, I don't know if that's my perception as I've not played for a couple months but flat out in the caliburn feels positively pedestrian now.
Click to expand...
agreed I fired it up last night and it still doesn't feel quite right.. At this point I think ill just finish it up and then call it a day
 
Rroff said:
Developers/publishers use all kinds of approaches, mostly BS, for this kind of thing - but game code just doesn't work like that. For instance the source of 2 games could be equal size but say one of them has dozens of monsters which all use variants of the same underlying code which is already robust and minimal chance for further bugs while the other has a complex feature such as perk mechanics that touches almost every part of the game using the same number of lines of code with the potential for 100s of bugs.
Click to expand...
Yeah but I’m suggesting you work with that. Specifically I’d bet that the number of bugs found by QA while initially all over the place decreases over time, towards an asymptote. You could use that to predict total bugs present and set your cut off point. It’s certainly better than assuming X bugs exist based on comparable games.

But hey, it’s still an estimate and as you say I don’t know the industry so my idea could be almost as wrong as the original.
 
You must log in or register to reply here.
Back
Top Bottom