Dolphin, the GameCube and Wii emulator - Forums

SMG runs a lot slower with LLE turned on for me, and that's a wii game.

About 30-40 vs. 50-55 if you want specifics.

Edit: By SMG I meant SMG2. And yes I did have LLE on thread turned on if you're going to ask me that.

I have no idea what is going on with rpglord's train of thought right now (as far as what he thinks Shonumi has been saying this whole time). It is clear that with framelimit off he is correct, just as Shonumi is correct (THEY BOTH THINK THAT FPS IS DIRECTLY RELATED TO POWER OF COMPUTER), but with framelimit on the difference between HLE and LLE on a low end would be much more noticeable because a high end system can do them both at full speed (which Shonumi supports but rpglord either refutes or has not stated an opinion, it is hard to tell).

(09-22-2012, 07:47 AM)NaturalViolence Wrote: [ -> ]SMG runs a lot slower with LLE turned on for me, and that's a wii game.

About 30-40 vs. 50-55 if you want specifics.

Edit: By SMG I meant SMG2. And yes I did have LLE on thread turned on if you're going to ask me that.

Ok,I just tested SMG2 as well !
Can you try with latest dolphin,vbeam on,AA disabled,idle skippin ON,and efb copies set to textures,and framelimit OFF
These are my results :
Starship : 60 vs 65 fps
Yoshi Star ( when you first land dont move ) : 64 vs 67 fps
Flufy Bluff : ( when you first land ) Same speed on both LLE and HLE : 52 fps

---

(09-22-2012, 07:55 AM)Axxer Wrote: [ -> ]but with framelimit on the difference between HLE and LLE on a low end would be much more noticeable because a high end system can do them both at full speed (which Shonumi supports but rpglord either refutes or has not stated an opinion, it is hard to tell).

Of course with framelimit on difference between HLE and LLE on a low end would be much more noticable.
But I stated when we begun discussion I am not going to test games which run fullspeed so I mention Last Story.
I also mention in a post after that I am testing with framelimit off.

(09-22-2012, 07:55 AM)Axxer Wrote: [ -> ]but with framelimit on the difference between HLE and LLE on a low end would be much more noticeable because a high end system can do them both at full speed (which Shonumi supports but rpglord either refutes or has not stated an opinion, it is hard to tell).

Of course with framelimit on difference between HLE and LLE on a low end would be much more noticable.
But I stated when we begun discussion I am not going to test games which run fullspeed so I mention Last Story.
I also mention in a post after that I am testing with framelimit off.

[/quote]

But you fail to recognize that Shonumi is saying something almost identical!

Shonumi is saying that when framelimit is off, the difference is the same, with it on, it is noticeably worse on a low-end system, and here you are saying the same thing!

</exclamations>

You are both saying the same exact thing. You just fail to understand that Shonumi agrees with you, but you can't seem to figure that out and continue to refute him/her, which makes Shonumi not understand what you are saying.

(09-22-2012, 08:10 AM)Axxer Wrote: [ -> ]You are both saying the same exact thing. You just fail to understand that Shonumi agrees with you, but you can't seem to figure that out and continue to refute him/her, which makes Shonumi not understand what you are saying.

No,he is saying difference,or "the gap" is not the same,even without framelimit.
Maybe he is right,if NaturalViolence gets 20 fps difference and I get 3-4 fps difference.
I am waiting for NaturalViolence results with same settings I used...

Shonumi Wrote:Look back to where this discussion began. It started when I claimed that HLE is much faster than LLE. That's the issue. You then claimed that HLE wasn't significantly faster than LLE in your experience. One reason I gave for your experience was that our hardware (which are fairly high-end for Dolphin, compared to most laptops and AMD systems) was sufficient enough to run LLE at or near fullspeed. If HLE runs fullspeed, and LLE runs at or near fullspeed, the difference isn't going to be terribly noticeable. That's why Axxer usually leaves it on LLE. A better way to examine the difference, as you noted, is to turn off framelimiting, then run both. Now in my tests, HLE sometimes runs twice as fast as LLE.

Sounds like he is saying the same as you from this comment.

(09-22-2012, 08:15 AM)Axxer Wrote: [ -> ]Sounds like he is saying the same as you from this comment.

(09-22-2012, 04:09 AM)Shonumi Wrote: [ -> ]We agreed that the gap doesn't just go away, not that it doesn't change. We're still at odds over what the size of the gap will be. You still seem to think that since the gap is small for you on your system, the same should be true for people with lesser hardware. I still maintain that on weaker hardware, this gap becomes more and more noticeable, at least until the hardware becomes so weak as to be unable to adequately play Dolphin at all.

I have clearly stated I am testing with framelimit off,and testing games which dont run fullspeed even on systems better then mine.
Why would he assume we are still talking about framelimit on,or only about games running fullspeed on my system ?

I have read through the thread now.

I can barely understand what rpglord is saying because most of his posts are a barely readable mess (no offense but you really need to improve your English). So I'm not going to bother responding to any specific posts in case I misinterpret something. Instead I'll just quickly discuss the subtopics that have been brought up.

The user experience (audio speed) a.k.a. Shonumi's point:
1. HLE is able to emulate audio at fullspeed on nearly any system and with nearly any game. It's that fast. Keep in mind I'm talking about audio speed here not game speed.
2. LLE on the other hand can only emulate audio at fullspeed or near fullspeed on extremely high end hardware for most games.
3. Users that are actually playing their games instead of just benchmarking them should have framelimit on because it's on by default and running a game faster than 100% is pointless unless you're benchmarking.
4. Therefore assuming you have framelimit on with a low end system there is going to be a tremendous difference between audio speed with HLE and LLE.
5. With a high end system audio speed is going to be 100% or nearly 100% with HLE or LLE.

The user experience (gamespeed) a.k.a. Shonumi's point from a different angle:
1. Gamespeed is a very different story. Gamespeed factors in a number of other things that are specific to your hardware and your game.
2. Once again if we are talking about the user experience we must assume that framelimit is on.
3. HLE runs asynchronously with the cpu thread and LLE runs synchronously with the cpu thread.
4. Therefore HLE can maintain good audio emulation speed even if the gamespeed is low.
5. LLE cannot. With LLE if gamespeed goes down audio emulation speed goes down.
6. With LLE on thread off LLE should have a big impact on gamespeed in almost every scenario. This is the default setting.
7. Therefore on low end hardware LLE is going to perform poorly.
8. With HLE you have a much better chance of running that game at fullspeed on low end hardware.
9. On high end hardware you will be able to maintain 100% gamespeed or nearly 100% gamespeed regardless of whether you're using HLE or LLE.
10. Therefore the difference is large on low end hardware and small on high end hardware.
11. With LLE on thread on the amount of performance decrease will depend on how much load is being placed on the emu thread (the cpu emulator).

The actual audio speed:
1. Not even worth discussing since we have no way to measure it.

The actual gamespeed:
1. This assumes measurement with framelimit off for benchmarking purposes.
2. With LLE on thread off HLE will definitely always be much faster in unlimited gamespeed.
3. With LLE on thread on the amount of performance decrease will depend on how much load is being placed on the emu thread (the cpu emulator).

Shonumi is correct no matter how you look at it ASSUMING that the user is using the default settings (LLE on thread off, framelimit on). I suppose it's possible for there to be a scenario where the performance difference between HLE and LLE is low with certain settings but I have yet to see someone report that. Everyone who has ever tested this and actually reported their findings has said that HLE was a lot faster for them. There is however a severe lack of people testing this with LLE on thread turned on so I will admit that until users do more testing with that option we don't really know for sure how that will effect things.

Edit: On second thought I should have just written this in paragraph form instead of list form. Oh well.

Shonumi Wrote:Oh boy...

Now you know what it's like to be me for a day. Horrifying isn't it?

Quote:Can you try with latest dolphin,vbeam on,AA disabled,idle skippin ON,and efb copies set to textures,and framelimit OFF

I had framelimit on but other than that those were the settings that I used. There is no point in using framelimit with this game on my hardware because it rarely reaches 60 fps.

This ^^ ( © by ED2 ) ( © by DefX )

Well,NV,if you say you get significant difference in SMG2 then I guess my theory was not correct ( regardless if Shonumi understood what I was talking about )
How do you explain NV,why exactly you seem to have 15-20 fps difference with same settings as I,and I only have maximum 5 fps difference ( when on starship )
Same load should be placed on LLE DSP since we are running same game and hopefully in same places in game.
Did you maybe test some other areas of the game ?
Or even more interesting question,if I dont get significant perfomance drop in SMG2 and you do,and thats is because I have faster cpu ( what else could it be ) then why I do get significant drop in gamecube games ????