• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Really, what is a "poorly-optimized" game?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
tviceman said:
Load up just about any Ubisoft game (sans far cry series) and that is your definition of poorly optimized. They are also notorious for showing off games at E3 with graphics way better than the final version ends up with (like Crysis 1).
Click to expand...

Crysis is Crytek developed and published by EA. Crytek left Ubisoft as their publisher after a dispute, and lost the rights to the Far Cry name in the process. Far Cry 1 was developed by Crytek and published by Ubisoft, but the rest are Ubisoft games.
 
nitromullet said:
Crysis is Crytek developed and published by EA. Crytek left Ubisoft as their publisher after a dispute, and lost the rights to the Far Cry name in the process. Far Cry 1 was developed by Crytek and published by Ubisoft, but the rest are Ubisoft games.
Click to expand...

You're right, but I think he's saying that FC2 and FC3 are actually pretty well-optimized on PC, despite being made by Ubisoft.
 
Morrowind, Oblivion, and Neverwinter Nights 2 are all quite badly optimized. Basically any RPG engine pre-2007...even today these games can chug at 25fps on a top end system with the view distance set to max.
 
Astrallite said:
Morrowind, Oblivion, and Neverwinter Nights 2 are all quite badly optimized. Basically any RPG engine pre-2007...even today these games can chug at 25fps on a top end system with the view distance set to max.
Click to expand...

This is where I take exception to the term, "badly optimized". While those games, if you turn every thing to max, can bring any system to their knees, they all run quite well using their predefined settings. They just happen to give you sliders to control every setting imaginable, allowing you to torture your system.

Just because sliders exist that can punish your system, does not make a game poorly optimized. Poorly optimized games, are games which run poorly on standard settings, and even low settings, due to poor coding. PC's are designed around allowing the user to optimize their own experience with settings given. You can't punish them for giving you options. Crysis 1 gets lumped into this by many people as well, because they had high end settings available.
 
Last edited:
and please lets NOT defend Crysis 1. it was most certainly poorly optimized. you could run custom configs that looked better while also running better at the same time.

even today that silly old game cant even stay above 60 fps unless you have a Haswell cpu. with a 2500k at 4.4 I dropped into 50s easily in some spots that were cpu limited. at the same time most parts of that game are insanely gpu limited even with most of today's cards.
 
toyota said:
and please lets NOT defend Crysis 1. it was most certainly poorly optimized. you could run custom configs that looked better while also running better at the same time.

even today that silly old game cant even stay above 60 fps unless you have a Haswell cpu. with a 2500k at 4.4 I dropped into 50s easily in some spots that were cpu limited. at the same time most parts of that game are insanely gpu limited even with most of today's cards.
Click to expand...

Yet, drop the settings to high, and even when it was released, it ran well. Granted, you can't get super high FPS, even today, but you can easily run at 60 FPS. Although there was/is a memory leak somewhere that is really bad towards the end.

I don't like to punish a game, just because they allowed you to use higher settings. If you do, they comeback with release day Crysis 2, where it was incredibly easy to max out. We had to wait a few months for high end settings to be released.

That said, it is true that they improved things in future engines, but that is the nature of progress. I just hate people calling a game with high end options, to be unoptimized.
 
Games that run like crap no matter how much hardware you throw at it are poorly optimized, because the bottleneck lies in the code itself and not in the hardware.

A few examples are Watch Dogs, Borderlands 2 (with PhysX), the Evil Within..

Watch Dogs is the latest AAA game I can think of that fits the bill perfectly. No matter how fast your CPU or GPU is, or how much VRAM or RAM you have, the game is guaranteed to run like crap with ultra textures because the engine's resource management is fundamentally broken for PC architecture. The fact that Ubisoft still hasn't released a patch to fix the problem is quite indicative of how deep it is..

As for Borderlands 2, they bit off more than they could chew with PhysX. From my own testing, the PhysX itself isn't poorly optimized when ran on a GPU. But the fact that the game is using DX9, the amount of extra particles and objects generated by PhysX being turned on overwhelms the single threaded renderer. So because of design choices, the game is tremendously draw call limited.
 
toyota said:
even today that silly old game cant even stay above 60 fps unless you have a Haswell cpu. with a 2500k at 4.4 I dropped into 50s easily in some spots that were cpu limited. at the same time most parts of that game are insanely gpu limited even with most of today's cards.
Click to expand...

I think thats mostly the part of the DX10 API. The API only used a single thread for rendering, which explains why you need a high IPC CPU to extract maximum performance out of it. If Crysis was remade using DX11, it would run much, much faster and with even higher visual fidelity.
 
bystander36 said:
This is where I take exception to the term, "badly optimized". While those games, if you turn every thing to max, can bring any system to their knees, they all run quite well using their predefined settings. They just happen to give you sliders to control every setting imaginable, allowing you to torture your system.

Just because sliders exist that can punish your system, does not make a game poorly optimized. Poorly optimized games, are games which run poorly on standard settings, and even low settings, due to poor coding. PC's are designed around allowing the user to optimize their own experience with settings given. You can't punish them for giving you options. Crysis 1 gets lumped into this by many people as well, because they had high end settings available.
Click to expand...

They are poorly optimized, as anybody who played any games prior to 2006 knows that back then programmers were just starting to implement occlusion culling in their engines as large 3D game worlds were starting and we were finally getting out of the "every 3D game is a corridor shooter" design.

Prior to aggressive occlusion culling being implemented in games, 3D engines would make the CPU do floating point calculations for objects off-screen which killed performance as you increased view (rendering) distance, because even though you only had say a 90 degrees field of view, 270 degrees (your blind spot) was being rendered and was a waste of resources because your system was doing 3-4 times more work than was needed to render the viewable frame.
 
Last edited:
AdamK47 said:
Try playing the Ukrainian game Cryostasis. It still stutters like crazy. Great story and atmosphere though.
Click to expand...

I think that was because the game only runs well on Nvidia cards. I had an AMD card at the time and it was a lost cause.
 
Astrallite said:
They are poorly optimized, as anybody who played any games prior to 2006 knows that back then programmers were just starting to implement occlusion culling in their engines as large 3D game worlds were starting and we were finally getting out of the "every 3D game is a corridor shooter" design.

Prior to aggressive occlusion culling being implemented in games, 3D engines would make the CPU do floating point calculations for objects off-screen which killed performance as you increased view (rendering) distance, because even though you only had say a 90 degrees field of view, 270 degrees (your blind spot) was being rendered and was a waste of resources because your system was doing 3-4 times more work than was needed to render the viewable frame.
Click to expand...

Based on that idea, all games before 2007 were poorly optimized.

Just because programming techniques have improved, doesn't make all older stuff poor.

That said, they may not have been greatly designed, but just because you can't turn draw distance to max, doesn't make the game poorly optimized. If they simply didn't give you the slider, would you then consider it good? After all, you could easily max out the game then. Just because settings exist that push the system too far does not make them poorly optimized.
 
My view is simple. I'd say any game that's released since 2010 and doesn't benefit from more than two cores/threads is badly optimised. Games that also are DX9 only could also have run better if they were using DX11 instead.
Granted, it's very much about financial costs, but that could be said about every project.


I think people are unfair to simply call The Evil Within a bad port though. That game is apparently running badly on the consoles as well, not just the PC version.
 
Last edited:
AtenRa said:
I dont believe this happens on Win 8.1. Take a look at Civilization V on Win 8.1Pro 64bit and FX8350.
Click to expand...
It's not that it definitely is or isn't happening, but without clear long-term pegging of a few cores, it's hard to determine with just the performance graphs (if the same data were stacked in the same graph, it would be a lot better).

If there is reason to keep more cores active, not doing so, and parking, will lead to lower performance than bouncing around. If there aren't enough time slices to keep all those cores very busy, then not using them will lead to higher performance. Making that decision correctly is not simple in the first place, doubly so with MT being common, and doubly so again with Turbo features, and then once again for the same scheduler trying to be good at making CPUs conserve battery power. The only scheduler I am aware of that comes close to doing it ideally (and which has commonly been used for Android) has a problem scaling much past ~10 logical cores (CK's BFS patches for Linux), and isn't stuck with Windows' compatibility concerns.

Win 8 is better than Win 7 at it, which was better than Vista, which was better than XP, which was better than 2K, which was better than NT4...but none are anywhere close to perfect, and won't be any time soon. Each iteration can get better, due to having more cores and cache on newer sockets to actually test against, where provisions for more were speculation-based in the prior OS version.
 
Last edited:
Generally when people complain about a poorly optimized console port they refer to the game code still balanced to the console strengths instead of PC. For example consoles can handle more draw calls than PC (with DX anyway). If they do a straight port you end up draw call bottlenecked. New consoles have a lot of VRAM available. More than the avg. PC now. Straight port will need more VRAM than the typical PC offers. We're seeing some of this now. They would have to redo the PC port to optimize it for the limits in these areas on PC. On the other hand the PC offers way more overall power if the game is optimized to take advantage. Stronger faster CPU's (within the limits of DX overhead) but they need to be coded to balance the performance across the cores. Same with the GPU's. You can add effects that would choke the console GPU (AO, AA, etc...). That's optimizing for the PC.
 
bystander36 said:
Based on that idea, all games before 2007 were poorly optimized.

Just because programming techniques have improved, doesn't make all older stuff poor.

That said, they may not have been greatly designed, but just because you can't turn draw distance to max, doesn't make the game poorly optimized. If they simply didn't give you the slider, would you then consider it good? After all, you could easily max out the game then. Just because settings exist that push the system too far does not make them poorly optimized.
Click to expand...

They were poorly optimized because Oblivion shipped with a version of the Gamebryo engine that had no occlusion culling even though the developers knew it was an issue that's plagued 3d RPGs including the prior game Morrowind but wanted to rush the game out. They patched the game with a new build of the engine a few months after release that completely fixed the blind-spot rendering issue, and IIRC was the first major RPG to implement it even though the techniques had been pretty widely implemented in shooters by 2004-2005. It's just RPG game makers are several years behind compared to other genres in programming techniques for whatever reason, maybe the scale of the games just means they overlook a lot of things like optimization on the programming side.
 
Last edited:
My favorite RTS games load the AI onto one core. Only way to get higher fps is to OC!

Sins of a Solar Empire: Rebellion
Supreme Commander: Forged Alliance

Both games are also 32 bit which means that if they hit 2gb of ram, they will crash. It can be fixed with Large Address Awareness to give it an extra gig.

ARMA 3 is another game that threads AI poorly.
 
Someone should put these same analyzing skills to Space Engineers. It's a fun little sandbox game, but it seems VERY poorly optimized for building lots of stuff... something about the graphics engine drawing everything regardless of if you can see it or not... or something like that. It bogs down FAST.
 
I intuitively think of poorly optimized as any game that runs much worse than you would expect, given how good it looks. This is often due to one or two graphics settings that don't do much visually but drop performance a lot (often token features tacked on to the PC version of a console port), but a few games run badly regardless of the settings or hardware.

AdamK47 said:
Try playing the Ukrainian game Cryostasis. It still stutters like crazy. Great story and atmosphere though.
Click to expand...

Try disabling hardware PhysX. I remember that hurting performance a lot for no reason (even though that was what the game was known for ironically). That game is on my to-play list with my 980, as it ran like a dog on my old card.
 
Almost any game that uses gpu physx. Watch Dogs is one of the most unoptimized piles I've ever tried to play. The old Stalker games were terrible for the time as well, but easy to run with today's hardware.
 
Astrallite said:
They were poorly optimized because Oblivion shipped with a version of the Gamebryo engine that had no occlusion culling even though the developers knew it was an issue that's plagued 3d RPGs including the prior game Morrowind but wanted to rush the game out. They patched the game with a new build of the engine a few months after release that completely fixed the blind-spot rendering issue, and IIRC was the first major RPG to implement it even though the techniques had been pretty widely implemented in shooters by 2004-2005. It's just RPG game makers are several years behind compared to other genres in programming techniques for whatever reason, maybe the scale of the games just means they overlook a lot of things like optimization on the programming side.
Click to expand...

I don't object to that problem.

I objected to you saying they were poorly optimized because you couldn't set the slider to max viewing distance. Games should not be punished for offering high end settings.

Though I'd have to see more evidence from reputable sources before I'll just take your word for the above.
 
bystander36 said:
I don't object to that problem.

I objected to you saying they were poorly optimized because you couldn't set the slider to max viewing distance. Games should not be punished for offering high end settings.

Though I'd have to see more evidence from reputable sources before I'll just take your word for the above.
Click to expand...

So this was a pointless conversation?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top