V 3.0 performance discussion

[Ssnake]

You're kidding, right?

I mean, yes I have 1Gb of video ram, but the card itself is just a cheap little GT610...

Well, it may or may not work for you - but it has cleared at least the 1GByte VRAM hurdle, that's a start. Also, keep in mind that I tested only two screen resolutions - 1280x720 and 1920x1080. And we didn't vary the Details settings, which still have a bit of an effect (even though we managed to improve the performance with respect to the ground clutter objects). So, by reducing the terrain detail, texturing and ground clutter you might be able to save enough performance for shadows. It also depends a bit on your personal sweet spot between image quality, screen resolution, and frame rate.

[Ssnake]

In the Level 5 Sticky, which Phenom CPU was used?

Combined with the GTX560? A Phenom II @ 3.31 GHz.

[pii]

I think it'll do fine.

1 GByte video ram should allow for shadow mapping level 3, maybe even 4.

After looking at the spec for the CPU and Graphics card I'd say he is going to have problems. That GPU is really low end, he would probably do better with a newer CPU that has graphics built in, that's how low end it is. Some of the newer CPU/GPU are getting surprising good. http://www.videocardbenchmark.net/gpu.php?gpu=GeForce+GT+610

I'm sure it runs better than what he had but my motto is if you're going to upgrade UPGRADE :clin: But then again while I have a nice system I'm broke. lol

[Scrapper_511]

Combined with the GTX560? A Phenom II @ 3.31 GHz.

Yes but there are several Phenom II's at that clock speed with varying numbers of cores and single-core performance; Phenom II X2 560, Phenom II X4 960, and Phenom II X6 1100T.

On another note, I just discovered that my Phenom X4 970 has very similar single-core performance with an I7 940. Which means the I7 940 & Radeon 7850 combo in the Level 4 sticky is a very close comparison to my rig. "Low Performance" rating here actually gives me a bit of relief.

[Maj.Hans]

After looking at the spec for the CPU and Graphics card I'd say he is going to have problems. That GPU is really low end, he would probably do better with a newer CPU that has graphics built in, that's how low end it is. Some of the newer CPU/GPU are getting surprising good. http://www.videocardbenchmark.net/gpu.php?gpu=GeForce+GT+610

I'm sure it runs better than what he had but my motto is if you're going to upgrade UPGRADE :clin: But then again while I have a nice system I'm broke. lol

My CPU itself should be just fine. I mean, eight cores running at 3.1Ghz isn't exactly top of the line, but it was the biggest one I could afford at the time and I had NO working computers at that time. The onboard audio on my old PC went out, and I took that as a sign to upgrade. I went shopping for a new box with a limited budget.

My video card is probably crap, and I'm not shocked. I went with pretty much the cheapest one they had, just so that I could have something a little bit better than the onboard...Er, actually I don't think I have an onboard video...I have nothing invested in it so I'm not hesitant to replace it. I think I honestly paid $15 for it.

[cobrabase]

This wait is getting old.

[Razor]

Yes, really old. I thought the delay was going to be a few weeks, not many months.

[Gibsonm]

Well this is what happens when people demand a release date (unrealistic or not) and are surprised when it slips.

I’m still working on “it will be ready when its ready“.

[Tacbat]

Cope.

[DarkLabor]

Shit happens even more in programming.

Just be patient guys, they just do their job.

[Ssnake]

We had an in-depth review of the situation yesterday. The new particle system turned out to be a lot more difficult to implement than expected, due to the ancient ways of the original implementation. So three weeks ago we decided - as a time-saving intermediate measure - to brush up the old (fixed) particle animations and to diversify them a bit, which however turned out to be more complicated (again) than expected ... due to the ancient ways of the original implementation.

I admit, this feels a bit like Groundhog Day.

Anyway, we'll fix up the last remaining bugs and what we can in the particle department in the remaining time. It is our firm intent not to spend another month with this.

[skipp]

Does this mean that

1. Version 3.0 is released with a brushed up version of the old particle system but the new particle system will be released at a later date.

Or

2. Version 3.0 is released with brushed up version of the old particle system but the new particle system will never be released due to the problems implementing it.

I'm hoping its the first one

[Gibsonm]

Well option 1 is open ended.

If they release the “improved particle system” as part of say SB Pro PE version 5.0 in 3yrs time, it would still be valid.

And again under your Option 2 “never” is a pretty dramatic statement.

Yet again just wait and find out.

[Captain_Colossus]

How about making the 'Destroyed' mod folder work- since the new damage system is going to be shelved or put on hold, it seems like an easy change that no one even has to use if they don't want.

[Gibsonm]

How about making the 'Destroyed' mod folder work- since the new damage system is going to be shelved or put on hold, it seems like an easy change that no one even has to use if they don't want.

Ah the ever popular “it seems like an easy change“.

I suspect that :

Anyway, we'll fix up the last remaining bugs and what we can in the particle department in the remaining time.

Is that whatever is easy to finish in the remaining time will be, and that’s it.

I suspect that any alleged “easy changes” may well be somewhere down the track.

[Captain_Colossus]

It's all relative of course, I'm not coder. But you would agree that it would be easier than the particle system- right?

The Mod system with folders is already there, you already have a mod system for skins, it's already in place. It's a matter of the program just accessing the Destroyed folder. It's not a new thing to add, more like something that's there which isn't turned on. Even for someone who doesn't program for a living, I don't believe that would be a major effort.

[Maj.Hans]

I'll seriously pay you $40 just for the new vehicles and civilian traffic.

All of the graphical stuff I consider to be nice to have, however I just want to point out that I bought two separate copies of the original Steel Beasts well after it was released. I'm here for the fire control systems and the physics of blowing up Russian tanks.

[Captain_Colossus]

With this bunch, you have a hobbyist group that has always sacrificed some degree of flair for the sake of what it makes up for in subject matter. I could still live with SB Pro's infantry sprites, they still did the job. But you couldn't go back to SB1's graphics, all other things being equal, right? After a few years of that, there was a case that that even a niche group such as this community still desired graphical upgrades- just willing to wait longer for these changes than people might playing Call of Duty or something.

I remember the discussions on the old SB site- people generally said the same thing, graphics weren't a priority, but there still was a growing voice for change as time went on.

Once the first screenshots of SB Pro Pe were popping up, you saw people getting excited about upgraded graphics, questioning whether the graphics shown were in their final state, so it could start sounding a little mythical if we always convince ourselves that we are such purists that we've always ignored a little more eye candy.

[Gibsonm]

It's all relative of course, I'm not coder. But you would agree that it would be easier than the particle system- right?

I have no idea.

If the guys have decided to forgo the new particle effects model “for the time being”* then I’m happy for any remaining loose ends to be cleaned up and the get it out the door (which I think is pretty much Ssnake’s intent).

Adding anything new at this point (“new“ = anything not on the current to do list) to me is out of scope and should be parked for a mid patch/update release before the next paid one.

Asking to add anything extra now starts you down the mission creep road, that they seem keen to avoid now.

* Duration TBA

[Captain_Colossus]

Asking to add anything extra now starts you down the mission creep road, that they seem keen to avoid now.

Not necessarily- you don't even know at what point eSim wants to wrap up the code, you don't know know how easy or difficult it is as you've just said. That's why I'm asking eSim of this, I'm not asking this of you.

If they do it cool, if not, it's cool too- but please, I'm asking eSim.

[Gibsonm]

I'm not asking this of you.

Alright but your post came straight after mine and contains the words ...

But you would agree

So I figured “I” was the “you” in question.

[Captain_Colossus]

You're right, you responded to what I was saying as if you were giving eSim's official 'no' at what I was asking of Ssnake.

I'm saying that I don't believe it would be relatively difficult to turn on a mod folder in lieu of spending lots of new time on a particle system. I would bet my life even in my ignorance about coding, that is to say, I think it's a rational assumption that doing one rather than the other would be relatively easier.

Ssnake can come back and say whether they think it's too low a priority or something, that would be totally fine, and that's who I'm asking. Maybe the question could be construed as being posed to the entire community, so for that I apologize that I wasn't more clear.

[Ssnake]

Because of the convoluted old implementation, we just HAVE to untangle the mess eventually. Therefore "not now" does not mean "never". It just means "not now".

At the same time the old particle effects are really worn out, and it would look weird to have shadows and normal maps and shiny specular highlights and all this ... and yet the same old smoke and fire that you saw in the last three to seven years.

There are limits to what can be done with the particles within the constraints of the old implementation, and there are certain complications even with the simplicity of the old system ... ach, it's no fun writing about it. I'm just saying, the waiting period is coming to an end. We're frustrated that we can't get the new engine to look as pretty as we thought we could make it, that there will be this obvious discrepancy in quality now - but a decision had to be made, and the alternative of waiting maybe another three months JUST for the particles was not really attractive either. Don't forget, it's not just about delayed earnings with SB Pro PE for us, it's also about the workplan for other projects for which we have contractual obligations. We have used up the time for the PE development including a generous safety margin (and a major shuffle within the work plan).

There will definitely be a visual improvement with some of the fire/smoke effects, make there no mistake! It's just not going to be more than 10% of what we had originally planned to implement.

[daskal]

@Ssnake: are there any WIP screens of the new (not yet implemented) effects that you could share with us in the meantime? Just to get an idea what we could expect for the future?

[Lt DeFault]

Don't forget, it's not just about delayed earnings with SB Pro PE for us, it's also about the workplan for other projects for which we have contractual obligations.

Do whatever you need to do to stay in business...I'll wait.