DX9 / XP discussion

[Blackdog_kt]

Originally Posted by Baron

I was only referring to the debate about "they MUST support DX9 cuz the box says so", if its a beta release they don't. They can hold of the DX9 support as long as they care to as long as its not an official/steam patch release. There is absolutely nothing that indicates that this patch is going to be a official release right of the bat.

I might have missed something though.

It seems like you are half-right, half-wrong:

1) Yes, a test, non-mandatory patch can skip a feature or two and focus on what it wants to test.
2) The final patch however cannot and must work towards the inclusion of all features stated/advertised.
3) These feature too have to be tested before release, with a test patch of their own.

That's what the next patch is probably going to be: finalize and test DX10 optimizations, test DX9 optimizations, test FM changes.

It's still going to be a test patch before it goes final.

I think the misunderstanding comes from the fact that many people treat alpha/beta patches as final ones and expect them to be gameplay changers like the final patch is supposed to be, when in fact they are meant to test the final patches. That means possibly doing things in a couple of different ways on each patch and watching for feedback to decide which is eventually best.

In that sense, saying DX9 users are holding us back is totally untrue. It's not them who are doing it. It's the choice of initial user requirements advertised and published that do so, because these requirements have to be satisfied:implemented, tested, finalised and released.

Originally Posted by MD_Marx

Thanks for the update B6.

I also understand the frustration about the DX9 support, and largely agree with all the commentators. However, I wonder if these purchasers of the game - albeit justified in bitching from time to time - consider the legal ramifications to 1C if they simply stopped any DX9 development? Right or wrong, the game was sold as being compatible with DX9 - and that's a legal, binding contract - as far as I understand it - so there really isn't much choice in the matter - DX9 support has to be provided. But we die-hard flight - sim customers do have a valid bone of contention with DX9 support, in that we all know that DX9 is long dead, and it is unreasonable that the justified satisfaction of the majority should be deferred for not just the minority, but for pseudo economic reasons as well.

So can I make suggestion? Accepting that DX9 development should continue, is it not possible for 2 versions of the game to be developed; the first being the one for DX10/11+ (the main development), the second (of lower priority) for DX9 ?

The advantage would be that this would remove the requirement for a single piece of software to satisfy 2 disparate operating systems. You could even limit the amount of DX9 support on the latter version, having as a simple objective, crash-free, reliable operation, based on an earlier version of the FM/DM etc., and simply let it die a death. After all, even XP systems will eventually become truly obsolete in the East, and as a business model, it seems fairly odd why 1C would want to waste funds in developing complex software for an outdated OS - that the vast majority of the customer-base has already moved away from. These guys' gripes are valid; we are still waiting for the promised land but are expected to wait further because of a mistaken promise that 1C made i.e. DX9 support.

It is all very well arguing that 2 versions will not work because of the complexity in maintaining the 2 versions, but you need to balance against that, the additional software complexity required to support DX9, 10 & 11 AND NOT UNDERMINE either! Since you have taken the decision to consider the DX9 question further, can I ask that you consider this approach as well?

Cheers,

Marx

One would need to effectively create a second game more or less, because you can't have separate versions going final: the game is on Steam, so the same final patch is applied to every installation. I think the approach you describe would end up being more time consuming overall.

Originally Posted by 41Sqn_Stormcrow

Personally I think it is not doable for a small team like MG. Also it should pose no big problem for the software as such (other games manage to have both at the same time).

I do not mind either that they finally go to implement some sort of DX9 compatibility. What I mind is that they keep the majority of the users unnecessarily waiting by insisting to publish this in one big package.

I think what they need to do is:

1. Release the final patch for DX10/11 users as soon as it is ready and later release a patch that adds DX9 compatibility.

2. Focuse on the stability of the code for DX9 users only. Doesn't need to have the greatest performance or eye candy as DX9 is only mentioned as minimum requirement for running the game, not for enjoying the game to its full potential.

I agree with the sequence you describe: optimize for the majority first, then optimize for the minimum requirements. Releasing the updates separately however is not possible because of Steam. If you release a final patch that doesn't support DX9, these people simply can't fly online. If they are among the people who make the mistake of letting Steam auto-update everything without asking, then they can't fly offline either because their game will be updated to a DX10 only patch.

That's what i guess is forcing them to release everything in one package. I'm not usually the "i told you so" guy, but the truth is that when some of us were questioning the inclusion of Steam in the whole deal (back in the day, before release) there were people calling us "crankly old luddites"
Well, what we have right now is an exact case of what i had in mind as a limitation that could be brought about by the steam platform.

With every bit of automation, you lose a bit of flexibility and direct control. That's how it is with all things. This community decided to go with automation, so now we don't have the flexibility for separate DX9 and DX10 final patches. It's as simple as that.

Finally, this:

Originally Posted by Skoshi Tiger

Put off releasing the DX9 patch????? The way I see it last patch significantly improved the game for us DX10 users. DX9er's have been waiting a long time for their promised improvements. Guys this is their time, and from what B6 says, DX10 users should see significant improvements as well.

Get a grip peeps!

The game has been semi-playable for me even from the beginning and very playable the last few months and i don't even run top of the line hardware (i7 920 @ 2.6Ghz, Ati 4890 1GB, just 3GB of RAM and win7 64 bit).

There are people who BOUGHT the game (like everyone else here) based on the posted system requirements and haven't been able to fly a quarter of the time i have. These are customers too and they are entitled their frame rates just as much as a guy with a quad SLI setup. Spending one's money on hardware upgrades doesn't entitle one to preferential treatment. Spending it on the game however entitles one to using it. It's as simple as that.

And finally, since they say performance will improve for all, what is the problem with it?



This whole thing is reminding me of the discussion i had in the previous update thread. I was talking about a proposed feature for bombers and someone told me to report in on the bug tracker to get results faster. In other words, report a bug when there is none, just to get my way faster.
I told him that a bug is a mistake in implementing something, when the feature is just a proposition in my head there is no bug.

Otherwise i could go around posting things like "bug 1255643: I want the option of taking the squadron's dog mascot in the plane with me when i fly but i can't. We know for a fact that many squads had a mascot, so what gives? Please implement it ASAP".

Can you all see the difference? Dressing up a request for a new feature as a bug, just because the word "bug" carries more urgency?

Well, that's exactly the problem in this community. Whenever something is about to get fixed, impatience takes over and we end up with band-aid fixes because of community pressure. But the problem with those is that all too often in programming, you will have to go back and re-do them (or have problems in the future) because you didn't account for the big picture when designing your solution.

I say let them properly finish everything so that it finally WORKS. You know, instead of getting a whiff of something we like around the corner and going amok, like "i don't care about other customers who are legitimate users of this software and deserve equal treatment to me, just GIMME NAOOOOOOO"

Originally Posted by Damixu

If I could give a word of advice to developer team in order to meet the frustration and demand of restless customers: Please consider to provide smaller and more frequent patches and upgrades to the game. This would give us a concrete feeling that we are progressing instead of what whe have now, long stagnant periods of uncertainity and speculation and frustration waiting for next patch to come.

BlackSix I would appreciate if you would pass this message to Luthier. I thank you in advance.

This i can agree with, for things that are relatively simple to fix.

"Dear community, this week we fix all the bombsight, but only the bombsights".

After a few days "A mini-patch will be up on Steam soon, correcting the reversed prop-pitch controls in the 109 and 110".

Things like that which are easy to isolate and test, of course we could have more frequent updates of.

I think the reason we don't see this is that they've been focused on effectively redoing a big chunk of the game itself (graphics engine) instead of smaller fixes on individual modules.

Originally Posted by Blackdog_kt

I agree with the sequence you describe: optimize for the majority first, then optimize for the minimum requirements. Releasing the updates separately however is not possible because of Steam. If you release a final patch that doesn't support DX9, these people simply can't fly online. If they are among the people who make the mistake of letting Steam auto-update everything without asking, then they can't fly offline either because their game will be updated to a DX10 only patch.

Releasing a pre-final patch as release version with DX10+ does not change anything right now to DX9 users. Either they can play it now with all the patches already released or they cannot. They'll have to wait anyhow.

However releasing the patch in two steps would allow those 85% of the users with DX10+ support benefit from the improvements. I cannot see one slight inconvenient for the DX9 users if they have to wait a couple of weeks longer just because the others would not have to wait.

[philip.ed]

Originally Posted by Blackdog_kt

It seems like you are half-right, half-wrong:

1) Yes, a test, non-mandatory patch can skip a feature or two and focus on what it wants to test.
2) The final patch however cannot and must work towards the inclusion of all features stated/advertised.
3) These feature too have to be tested before release, with a test patch of their own.

That's what the next patch is probably going to be: finalize and test DX10 optimizations, test DX9 optimizations, test FM changes.

It's still going to be a test patch before it goes final.

I think the misunderstanding comes from the fact that many people treat alpha/beta patches as final ones and expect them to be gameplay changers like the final patch is supposed to be, when in fact they are meant to test the final patches. That means possibly doing things in a couple of different ways on each patch and watching for feedback to decide which is eventually best.

In that sense, saying DX9 users are holding us back is totally untrue. It's not them who are doing it. It's the choice of initial user requirements advertised and published that do so, because these requirements have to be satisfied:implemented, tested, finalised and released.

Blackdog, you can appear to be quite condescending at times. This is called wasting ten minutes of your life to regurgitate the poster you were quoting from. At no point did Baron indicate that DX-9 users were holding the patch back. In fact, releasing the DX-10 patch for testing would suggest a beneficial move forward in clamping down DX-10 issues, thus being a positive move forward (especially on the premise that the team's 'minor issues' tend to take weeks to sort out)
The fact that the team already released a Beta patch just for DX-10 users negates any suggestion that the two can't be released (in Beta form) separately.

You're right that all the features will need to be tested together