Marks Latest Worklog

Ok, have at it http://www.blitzbasic.com/logs/userlog.php?user=1&log=1043

Sounds like an unexpected but realistic move. Best of luck with Blitz3D2/3DSDK Mark.

hmmm... Best of luck! But it doesn't seem that there is a definitive choice made from reading that worklog entry. I'll just pretend it has never been written, see if I'll get a surprise for 2009.

Personally, I'm waiting for a release of Pure Basic 4.20. I've heard it'll have more support for Ogre 3D and Direct X 9, support for tracker modules from ModPlug, and a lot of other niceties. There's a lot of things in PureBasic that I don't like, but there are many many things I wish Blitz3D would have that PB does have.
http://www.purebasic.fr/english/viewtopic.php?t=32235

I was really tempted to go Cobra last year, but it's not for me. I just don't understand why the language and 3D module are sold separately. It brings the cost up, and the language is not fully matured imho.

Blitz3D is really a nice language, but it lacks a lot of stuff.... If only there could be a cross between Blitz3D and BlitzMax! Anyhow, I won't have any expectations for this, but it sounds interesting.

Makes me think of this post

I figured he would move away from BlitzMax as the platform for his 3d engine, it's simply not on the right road for that.

Why isn't it the right road?

Because writing it in c++ will produce a quicker 3d engine, and he'll be able to sell it to a wider audience not just BlitzMax users.

Good luck Mark!

But it's not clear as if wether it will be a 3D engine on it's own, or a total package (language + 3D engine) just like Blitz3D. Well he said it might be like a Blitz3D 2.0 but I'm skeptical on the outcome of this... I don't want to learn BlitzMax and a new 3D engine. Hopefully, it will be simpler than this.

Also, just curious, can BlitzMax make really tiny executables, like 4K or 10K ??? I think the smallest BlitzMax executable is like 400K, no?

I got 297k with a fairly large project (irc, and it links with razor - ultima online stuff), it has its own icon and an xp theme manifest. With BRUTE upx I get 95.5k.

EDIT: Definitely not, this:

SuperStrict

Framework brl.standardio

Print "Hello World!"
End

Compiles to 45.5k.

EDIT: Ha! another one that uses more code then the above example compiles to 24.0k with upx I can get 12.5k.

Sounds good to me.

Exciting news, thank you Mark.

From the worklog, it seems to me that both Blitz3D 2.0(3d engine) and b3d-sdk (library) will be cross-platform. We could continue to use Blitz3D commands, while b3d-sdk is for various languages.

Gives wide-eyed look!

mmmm, I dont know if anyone else felt the same, but I get the underlying feeling that Mark has ran out of steam.

That post has no confidence in it what so ever:-







Theres nothing in there that makes me wanna go 'ohhhh' and 'ahhhh', its, well, dire!

Best of luck with it though, but I doubt I'll be holding my breath for it!

That does sound very good indeed.

Dabhand, wait! The guy created BlitzBasic from the Amiga up to Windows, created a 3D engine, created BlitzMax, Maplet, etc etc, and you're doubting he can create a new 3D engine / Language ??? C'mon! Plus you haven't seen what he's got in the backburner.

But you know what, it's good and healthy to be sceptic.

Very good news :) But i just see for the moment another method to make me
pay , again and again. Because i doubt that the product will be free...

Because for sure, i'll buy it! I like give my money to BRL :)

> That post has no confidence in it what so ever:-

Ha! Just re-read it and you're quite right...shoulda used more CAPS and BANGS!!!!!

The truth is, I am quite disappointed the game creator thing didn't work out - I had 1001 ideas for it, but the whole thing just kept getting bigger and bigger yet more and more unclear. I was bouncing ideas off a few people via email, but I personally find that to be a really ssslllooowww way to develop ideas - you're always left with a feeling like 'Damnit! I'm sure if we were just in the same room together with a piece of paper we could nail this!'. So, my next attempt at game creator creation is likely to involve some real live programmers.

I also feel a little embarrassed too in a way - I've come more or less full circle and have sort of ended up doing what a bunch of people were telling me to do in the first place!

These 2 points were probably primarily responsible for the muted tone of the post...

But I've been hacking away at this over the last few weeks and have made good progress. I have by now tried out quite a few approaches(!) so have some idea of the 'best' way to go. Of course, there is no objective 'best' with this sort of stuff, but in this case my aim is really just to recreate the ease of use+flexibility of b3d, but with modern goodies.

Fortunately, I just happen to have tons of 'modern goodies' code lying around by now too! Over the past few years, I've built up quite a collection of code for doing physics, stencil shadows, shadow maps, decent shader interfaces etc etc, so it's hopefully just a matter of putting it all together into a sane library.

OMG, Filax, did you just mention the word free? :)

I think we need to get a list of features we'd like to be implememnted in the new Engine. As I suck really bad at 3D I'll leave it to you guys to state want you want to see in the new 3d engine. :)

e.g

New improved B3D file format.

Shaders for for both opengl and HLSL

Built in shaders to remove the need to have to program them ones self. i.e. Bump Mapping, Glow, Toon Shader, Glass, Water Shader, Screen Space Ambient Occlusion etc.

Fast Shadows, Shadowmapping Support.

Physics support like 3impact would be awesome.

Well, have fun doing it either way. No point doing something if you don't like it.

No...

Don't get your hopes up guys, just wait and see what pans out. You never know this time in 2years with no release mark could've posted another worklog saying he again didn't feel he could complete this b3d2.0 etc thingy and has moved onto another idea.

He's like a newbie coder moving from one idea/project to another, never getting it finished.

ps skeptical as i am, i do hope he does get something created that does justice to the blitz name.

As long as its a cross platform and can do what b3d can do, ill buy it happily.

Thanks for the information. Somehow i'm not surprised by it. To me it sounds after a possibly larger market but also less comfortable integration into BlitzMax and somehow less amazing. Curious to see how it will look like in the end.

i like the idea very much. wishing mark all the best!!

A bit disappointed by the worklog. The are usually much larger and contains lots of interesting stuff about how bmax works. :/

3D engine by Christmas? Sounds nice.
Good luck!

I really like the idea. I think the game creators can be Max itself, not exactly a drag & drop development environment, but mergin max with a good next gen 3d engine will convert max in a super sexy cross platform development tool.

I hope there will be a new version of Maplet and most of the existing b3d codes could be re-used.

Could you involve us a little more when your programming this? If you don't want to spark off mass hysteria of forums posts then that's understandable. But once a month work log posts would be lovely.

I agree. Please fill us in a little bit more often. Even if it's just short ones once a month.

erm.... anyways, looks like the powers that be at BRL do read whats going on in the forums. Just takes a lot of moaning to get them to respond :)

Amon : No :) i don't want hurt anybody! , but sometime, i'm feeling like a
pigeon...

I bought blitzbasic just before BRL stop the distribution for blitz2D, then
i bought blitz2D just before Blitzplus... And next, i bought Blitz3D and
Blitmax, and i'm waiting since 4 years a decent 3D engine (due to the max3D
screenshot,from Mark worklog) ...

And, i forget! i bought Blitz3DSDK too, to see it evolve, but, nothing...

When i hear : New "3D engine" and why not a "Blitz3D 2.0" i'm affraid for my
purse :)

Wait and see...

No!! Mark!! Put the gun down please :) Pan !

This sounds like very good news to me. BlitzMax 3D engine and/or Blitz3D_2.0 would make me happy. I still get more use from (and get more jobs using) Blitz3D than I do BlitzMax, but love them both.

I'm a great fan of B3D and bought Max too but never got on with it so this is really quite exciting news, I've been waiting for a Blitz 2.0 type all in one package without needing external libs/modules etc etc

Thanks Mark for finally coming to your senses.

boiled, he seems to be talking about an SDK. (For BMax, C++ whatever) not another "b3d"

I dont get what people keep going on about b3d 2? Like mark is going to just drop a great language like Max!
And besides Max with a naitive 3D module would be B3D 2! as far as most users would be concerned.

How would max with a native 3d module be so bad? (except for the fact it is not engine dependent, which is a bad thing anyway and is a root of many b3d probs)

Also, mark mentions working on his lonely self, well obviously he needs to get some more people on his side to help with his advanced projects cos no matter how good he is, what he is trying to do is too much work for one man.

Maybe he should get the likes of simonh and klepto, who are doing a great job on minib3d versions, working along side him and im guessing skidmarks too.

ps Get puki on your marketing team while your at it, he should get the word across nice and loudly :-)

Me either, Mark was quite clear. He is working on a new high end 3D engine usable with a variety languages (a new 3D SDK). Blitz3D 2.0 is only mentioned as a possibility and Mark gave no promise for it.

I agree that a regular update would be a good start too.

I would love to see a Blitz3D 2.0 come to fruition so all the best with that. Hopefully you`ll keep the 2D side of things in aswell, like in the original Blitz3D, as this was one of the main reasons that I was drawn to the language (2D and 3D all in one). Of course with realtime scaling\rotation etc... And make sure we can turn off that damned texture filtering this time :)

I hope that this works out for you Mark.

Jason.

I wonder if part of the reason for this change of priority is the plethora of posts in these forums asking for this very thing. I am very excited to get stuck into a B3D2.0 with updated graphics, (hopefully) cross platform as I and most of my friends are on mac, a performance boost and some things like unified automatic lighting/shadows, render-to-texture, decent GUI... I'd buy that for a dollar!

Blitz3D: Steroid Edition.

I'll take that for xmas!

Good call Mark, can't wait.

..well..while I am carefully controlling my excitement, due previous 'dragging', I must admit that Blitz3D v.2 making my heartbeat faster..if this thing appear, I am going to buy it as well as SDK version for use with other languages..all in all, i think its good news, but lets be patient.. :)

So would I if it had ALL the features of blitzmax... Oh wait, why not just use bmax which (when used at a basic level) is almost the same as b3d?

haha! he didnt mention WHAT christmas! even worse: he says a christmas, not even "this" or "the upcoming" christmas, but: "a"...like..."any"! another genius move by Mr.S. this is, how hypes work ;D

I am dissapointed with the worklog BUT please remember, the company is called Blitz RESEARCH. If Mark didn't take these routes, he wouldnt actually be researching.

Without trying an alternative route, how would you have ever known if they were correct?

Coming to your own conclusions through experimentation and experience rather than blindly following the throng should *never* be the cause of embarrassment.

Besides, I very much doubt the time was 'wasted'.


Now you've found your feet...Get bleedin move on! ;o)

It's a shame the engine won't be written in BlitzMax, but it's great to see Mark is 'back on the move'. Now the B3D and Max camps will both be happier (cue Churchill all of the people some of the time speech).

Oh... and to the brigade lead by Puki... I said you would never see B3D 2.0... I have eaten my hat.

I'm pretty burnt out on Blitz3d after finishing Tank Universal.

But there's something about the words 'Blitz3D 2.0' that's very exciting!

This is the best worklog update I've read from Mark!

Well I'm excited!!!
I still haven't released my first project, but this is great news!

Best of luck Mark.

I hope Blitz3d2 will be nearly identical to blitz3d in the basic stuff, so you could transfer an old blitz3d game with ease to anext-gen graphical level.

A B3D 2.0 sdk/library is great news for the forums - nice one Mark!

IPete2.

Good news.

I just read through the thread and i'm a bit confused if i got things right here. Is this about a crossplatform 3d/physics engine for BlitzMax plus other languages AND/OR about a Blitz3Dv2? Depending on how it looks and feels, i could be interested in the first option, not in the second one.

It is about a 3D Engine/Lib for BM and other languages and there is a "possibility" that B3D 2.0 might appear.

Some have jumped on the B3D 2.0 as it is set in stone even though Mark distinctly said it was only a "possibility".

Okay...thanks!

Smart move!

Blitz3D 2.0 is the way to go: give us a all-in-one package with an easy and fast language and a robust 3D engine: That's your selling point.

If mark is still lurking here:

Will the max3d module be sold seperatly from bmax?

Also, keep up the good work!

The problem with BlitzMax and a separate 3D engine is that it's too complicated for the common mortal. The Blitz3D formula is maybe a bit simplistic, but it works.

Since the current 3D engine for BMax is sold by itself, I would guess the new engine would be, too.

It's nice to see Mark talking about ease of use and accessability again rather than new programming methodologies or technological treats. The original appeal for me (and I guess most users) with Blitz was that it's simple, easy to learn/use and powerful. This was lost with BlitzMax and IMHO development has been mired "in the wilderness" ever since.

I hope the new development goes well and we see a next generation of great new Blitz products, although I hope it gets to market before the market dissapears!

Darkheart

I'm happy if we do get blitz3d2.0, but I'm not sure why. Will adding a next gen engine to blitz really mean I'll make the next gta?

I'm happy with it. I'm looking forward to working with the B3DSDK again. I was about to give up on it. (Then again, I have not done any coding for fun in over a year anyway)

Puki - I loved the 2 dog pictures...

No, unless you can produce the same content for your game that they did to gta.


How similar might the next 3D engine be to the current SDK?
Is it worth buying the SDK and start making a game and then switch over to the new engine when it comes. WIth none or just some small changes to the bmax code?

It really isnt. Anyway, a naitive 3d engine would be fairly seamless and well supported (like the way "max2d" works).

Alain,

Thank you for putting it so succinctly - that's exactly what I yearn for, what he said!

IPete2.

I welcome the prospect of a Blitz3d 2.0 but I bet we'll still be waiting for it two years down the line.

Basically if it can load Blitz3d files and run them off the bat then I will buy it, not the Max route where you have to basically do a recode to get big chunks of code to work.

Will most likely be the Max route. Since Max is the current language.

Will it be a Direct X engine or an OpenGL? If it's an OpenGL, it will be dissed by Windows Vista, if it's a Direct X, it won't be portable like BlitzMax is. Either side, you're doomed with BlitzMax (AND 3D). Sorry, I'm a bit nihilistic.

@IPete2, alain etc.
I dissagree totally and i dont see the point of making another closed embedded language which in a few years will be in the same boat as b3d is now.

Please tell me what you think an embedded engine would give you over max + module. If its "ease of use" then im totally confused. How hard is it to unpack a zip file into a folder?

BlitzMax has virtually the same basic command set of B3D, it has much better 2D commands and -if- the well known and loved b3D-style 3D commands were seamlessly added then it would be EVERYTHING B3D is/was and a whole lot more.

Absolutely no contest as far as im concerned.

I have been converting a lot of my code to MAX from b3d and aside from the 3D stuff, for the rest of it the only big difference (apart from a few very minor syntactical differences - mostly fixable with s&r) is the use of the TList when using types (and is better for it by the way)
- So. You can still program in a basic way like in B3D if you dont want to learn all the oop & more advanced stuff, so whats the problem?

@_33
Why? Support both APIs, OpenGL on OSX and Linux and DirectX/OpenGL on Windows.

@_33 OpenGL works fine on Vista.

I just hope that you do get a nice and easy commandset (like in Director for instance) with an underlying reasonable concept and a precompiled module which you just plug into a folder but without the need to install a compiler first and compile some c++ source on osx/win/linux.

Well for some reason I heard some stories about OpenGL and Vista where their OpenGL games didn't run. Apparently the OpenGL API in Vista were lacking from the graphics drivers that were installed. It is possible that this has changed since then.

stuff: http://www.opengl.org/pipeline/article/vol003_7/

Wow, what a turn of events... My take on it:

THANK YOU!!!!!


Repeating past mistakes again...?
You should know better than to say something like that.....

Getting back to the roots sounds wonderful however it might be a tad bit late, who knows.

Bottom line is there is already a track record set forth that will be hard to break. Sure, your only one guy, and sure your intentions are always in a good place ( no one can judge that but yourself ) however the fact that A. Deliveries have always been late IF at all and B. The communication, or lack of, just allows for more skepticism and C. the constant change of direction is very unsettling.

Sinu, in my opinion, sums up what I am thinking. Those with at least half a wit of common sense are thinking the same. And in that this would probably be considered your last shot at winning the hearts of the indie game creation community as only the true die hards have stayed to see the final show. My recommendation would be to work your arse off and turn over that new leaf. Sure it's going to be hard, and you'll get a lot of doubters in the process, however if you pull it off you will regain much of the respect that has been lost.

With the number of trials, I would think that you can create something that can be a nice blend of everything. Blitz3D, excellent product in it's time. Allowed you create some great stuff!. Blitzmax, much more object oriented. Blitzplus, GUI friendly! Perhaps a combination of the good features of each without going to far from the goal. Personally I got sick of using Blitz3D because trying to emulate OO was getting painful. If you could lean a bit more towards Blitzmax without going too far into it that would be great. Blitzmax for me was way to over-complicated, although a lot of that could be due to the poor documentation, but alas.

Lastly, if you need help with something, ask a few people. You have a community of programmers at your fingertips, use them. You don't have to give up the whole product source to let someone else write a particle library.

Then again, those are just my thoughts.....

No offense but director sucked sooooo bad....

I went from blitz->Director->c#->XNA->Unity
and I must say that director was a very painful experience. I am rather hung up on Unity at the moment however the fact that I need to have a beefy mac lying around (at no cheap price) is very discouraging.

I don't care what happens as long as it's cross-platform. :)

The OpenGL API is not a part of Vista (or any other version of Windows). OpenGL support is provided via the drivers put out by graphics card manufacturers.

@Dark Half
You have to differ. It's not the commandset or the concept which sucks in Director, it's that the 3d engine never got updated in all those years. Intel's shockwave 3d engine was brilliant for a V1 and were you can compare it the commandset often is easier than what Unity offers for instance, take for instance "clone" vs. "Instantiate", it's not case sensitive, less to write, just better thought out in a whole. But hey i also do favour a "group" over an "entity". :O)

I think there is some basic OpenGL drivers that comes with windows XP/vista made by microsoft. Those will be slow. It have been like that in the past at least. DirectX games worked ok, opengl was software emulated.
You got hardware accelerated OpenGL once you installed some drivers from the card manufacture.

yeah, the whole 3d thing was a bit of an after thought and just 'hacked' in imo.

Although I like how Unity allows you to use javascript and/or C#. Plus allowing you to make variables accessible by the editor by globalizing them in the script.

I don't like the languages supported in Unity at all. I would favor languages like Lingo, BlitzMax or Ruby instead. Languages which look less after some robot language and were you can type less for the same effect.

Of course, it would be illegal for MS to do otherwise. Depending on the version of Windows, the software emulation for OpenGL uses GDI or DirectDraw.

I do not own B3D or the 3D SDK, but I think Mark has come up with a good game plan. But no matter what he does, he will not please the majority of those in this community and they will gripe their bottoms off and put him down no matter what.

Just make the bloody thing scaleable! and you shouldn't bother making the physics for it, Newton would be muchos better.

I doubt he will code Maplet again, didn't Mark lose the code?

Why would newton be better than the code Mark have? Have you seen it?

my dreams comes true .
mfg panno

Great to have some news - especially good news. All the best with this latest project.

@Plash:

Even Mark has lost the code of Maplet, he could easily re-create a better and newer one, I believe! Agreed?

No I have not, I assumed it was unfinished. Thats why I suggested an easier route if it was the case.

Agreed! If he decides to continue 'the game creator' (max3d?) it will probably be integrated in that.

It all sounds good to me, keep us posted please.

The key to success

Sounds good to me, kind of like B3D 2 could be a game dev framework API with helper functions, like what we eventualy want EasyFlow to be. Something coded in C++ that can be used with any language has some big plusses including portability to consoles.

I see not sticking to Blitzmax for the underlying engine a big plus, even if it does mean closed source like B3D.

@deps newton is sweet and newton 2.0 looks to be even more awesome. Mark isn't a fullon physics guy like the guy doing newton. He knows what he's doing when it comes to a physics engine and mark should intergrate it.

In fact mark should just wrap aload of stuff ie physics, graphics etc into a blitz3d like wrapper, that would cut his dev time down a ton and it would be pretty sweet!

Is newton the new Ode? Some time ago people was really pushing for Ode and told others that Newton and Tomahak(orwhatsitsname) wasn't worth it.

Been ages since I used physics libraries, and it was that Tomahak one and I forced it to use only 2D since I wanted some boxes in a game. :)

**** Compile 2D games/multimedia to Flash would have major benefits and attract a massive new audience from the web developer community ****

A request from the Red October:-

http://www.blitzbasic.com/Community/posts.php?topic=60666

Seems like he was onto something there! :P

So where exactly does this leave BlitzMax? Should that now be considered a finished product which will never get the additions it badly needs?

And what about Mac users? I was seriously considering buying a Mac come the end of the year, but BlitzMax hasn't seen much action in quite some time, and now there's a "next-gen 3d engine" on the way, you have to assume that it won't be Mac-compatible or it would surely have been called a "cross-platform next-gen 3d engine".

I think the new Blitz3D SDK will most likely fit in well with BlitzMax. Support for exporting to the web should be incorporated into BlitzMax - this would be a big bonus for both game and web developers.

What does this mean then...?



I wonder?

IPete2.

Ah, I misread that bit. I thought he meant in terms of languages supported there, but rereading, probably not. Well that's something for Mac users at least.

That means there will be a 3d engine written in C++ and usable with C#/C/C++/B3D/BlitzPlus (probably)/BlitzMax etc..

I think alot of you are confusing the relation between this and b3dsdk, its probably going to be as simple a command set as b3d/sdk but be more modern.

I disagree with his words, especially with the DBPro bit. The reason I ended up buying BMax was because the DBPro demo would not run on any of my systems when installed, it would repeatedly crash. Cobra and its "Pure 2D" was so damn slow on a very highend system that it was completely unusable. BMax just worked and it has been rock solid and perfect for the apps I need, and the OOP is great.

I don't own any of the other BRL products, but based on my experience with BMax, I am confident that whatever Mark turns out will be great. I do not have any interest in 3D, or games for that matter, but Mark seems to really know his stuff when it comes to making a decent programming language.

Just my view,

Gary

I'm sure it will, but I don't really need 3D. I have TV3D, and if I wanted anything else, there is always Irrlicht, or Ogre. I'm thinking mainly about the language itself. I currently have this vision of BlitzMax rapidly becoming the next BlitzPlus.

Thanks for the news.
But what Christmas? [ironic...:)]

...I have tons of questions (and I think many of us) about the 3d library/engine (what are the min hardware specs, there will be some restrictions, pipeline integration, and so on...) but I hope we'll read the answers into the worklog.

This is good news, cant wait till december ... the b3dsdk-like library would be useful with PowerBasic.

I'll get excited when I see it. I'm used to seeing too many release dates and no releases.

I don't know what marks budget is, but it would be nice if he hired someone to update his old projects (b3d/max) since he seems to have no interest in them anymore. I would pay pay pay for core update update update =(

Max and b3d are capable but when things aren't ever updated core-wise by mark himself I feel that they are a dying breed.

For me, the statements in Mark's latest worklog sounds lik "Game Over for BlitzMax" for me, as it has implications that do not concur with my plans. But I appreciate the honesty and clear words - it saves me from a disaster.

I couldn't care less for a 3D game engine, but I really care for a robust multi-purpose language that makes life easier for me. This language now obviously became a step-child, and that's a clear no-go for a long term NON-GAME project for which I was seriously considering BlitzMax.

I really, really like BlitzMax, but after this statement, I know that I have to follow a different path (most likely Python for this project).

I safely assume one of these languages will be BlitzMax?

I don't see that at all. The following line from Mark definitely implies that the new 3D SDK will come out initially for cross-platform languages, e.g. BlitzMax.

Also, he clearly stated the reason for picking C++ is one of speed (i.e. not functionality), so I wouldn't say it means "Game Over for Max" at all.

As for Blitz3D2.0, Mark is clearly going to concentrate on the engine and SDK first, for use with any language, almost certainly including BlitzMax. Blitz3D 2.0 will (if at all) follow on afterwards, i.e. puki WILL have to wait until 2009 for it (if ever). In my opinion, a Blitz3D2.0 would actually be some kind of slimmed down BlitzMax + b3d2sdk + IDE anyway.

Now, just need to convince Mark to a) ensure there's a strong focus on good docs, b) not write yet another IDE for it, and c) post worklogs at least once a month. Focus on what you're good at mate!

Well, it would be a bit dumb if it wasnt :D

Precisely. Guess which cross-platform languages Mark is most likely to use to build some test apps that he can use on any platform...C++ first no doubt, but BlitzMax makes a great secondary (or even primary) test harness.

Blitz3D has been updated quite a few times since it's official release, for a number of years. I don't see why BlitzMax wouldn't get future support.

But, I wonder how the new engine will differ from Leadwerks Engine 2.0 ?!

The last release was "mainly a 'cleanup' release", that was 5 months ago.

The last real release was over 7 months.

About half a year, I'd say BlitzMax could do with an update.

Other than to roll-up SVN tweaks like the last release, what precisely would be in that update? And don't say threading because it's clearly "A Big Job(TM)".

I think BlitzMax overall is a pretty good product. It sure has it's lows like the docu, no threading support, no 2d physics, not so powerful list handling, the poor ide and i think there wasn't a lot of maintenance the last months but overall it's still fine. I just hope that it will get some further love in the near future.

From the 3d perspective an easy to use max3d module which is crossplatform useful and stable would be very nice to have. A pure Blitz3DV2 doesn't make any sense to me. I think this could be interesting for those who just want to play around with new 3d features. I'm quite confident that you will see dozens of badly lit normal map demos then. :O) But for those who want to use a 3d engine in a more reasonable way it doesn't sound appealing.

I think those are more interested in a nice BlitzMax 3D solution or they already went torwards Unity because the product already exists, at least for the Mac. What makes my alarm bells ring a bit is that i don't think a good engine is just done by glueing some code snippets/features together as it doesn't sound after a nice concept behind, the different parts might not fit together very well and cause ugly interactions, such things. Hard to say as you don't kow what he really has done but i just want to say that some loose tiles aren't automatically a well designed 3d engine. One advantage he seems to have is that he has walked on quite some roads already.

By the way i think the term next gen is quite misused as i don't expect things like realtime radiosity, softbodies, ai moved characters, ... in this engine. From the features i've read this sounds more after current and yesterday gen.

Well, let's keep on waiting and i hope it will turn out to something good in the end as i would be happy finally getting max3D for BlitzMax and it also will be interesting to see what others offer at the time this will happen.

Since the man itself already covered everything else in his post (physics, stencil shadows, shadow maps, decent shader interfaces)...

I have just one request for Mark Sibly (very important):
This time either give us support for some major modellers or an interchangeable format (.fbx/collada), or give us from the beginning a converter from one of the aforementioned formats to your new proprietary format! :)

I said more about it here:
http://www.blitzbasic.co.nz/Community/posts.php?topic=73251#861239




By supporting other chipsets besides ATI and NVidia high-end GPU's and having lower hardware requirements? I'm not critizing Leadwerks engine, I just think it aims at a different target.

EDIT: Never mind. I may need this breath one day.

It's a shame Mark isn't aiming to do anything to tackle these issues. Actually, the speed is good enough, but easier comms with the outside world would really help.

You wouldnt be the only one Gabs mate... Recent conversions I've had over the last 6 months is that people are getting weary.

We have also found that Mark works alone, granted Skid and Simonh (And not forgetting Brucey) work on the language in their own way, but it seems the sheer amount of development is from the man himself... That raises questions wether he can keep going with Max with updates AND bring a multi-platform version of Blitz3D to the table...

I love Blitz, but at the end of the day, Mark isnt a God or some magic coder... He's human and dropped the ball with Max3D, hopefully, he'll pick it back up and crack on with Blitz3D XR2 (And keep Max going), simply for the sake of the future of the language and the community.

Mark nowhere mentions multi-platform support for the 3D engine, only support for multiple languages. And still, I just don't care if that 3D engine will ever see the light of day or if it will be available for every operating system that was ever shipped.

I only need a supported programming language for my platform of choice. Egoistic? Sure. But in the end of this year, my plan is not to extend my current contract but finally start my own business instead and go full time into developing my own product. So at least the timing of Mark's post was right for me.

I would have preferred multi-threading support in BlitzMax over ANY graphics. Also, wxMax with BRL's official support would make more sense for me than any graphics engine ever could. And so would an official database API. And while we're at it, 64 Bit CPUs have become mainstream, so having a compiler for 64 Bit architectures probably won't hurt anyone. BlitzMax could be the killer development environment for OS X and probably even for GNU/Linux.

But even without going for big things like that: A really working MaxGUI would be nice as well. Anybody here ever tried to print something on OS X?

I know that I want something that BlitzMax officially never was. The funny thing is that I am not alone in this boat and that there are some others here that also use BlitzMax for tasks that are not related to gaming. It's evidence that the language is very good and has potential.

My problem is that I've now entirely lost my confidence in the viability of the product for --MY-- purposes. If the creator of the language prefers to create 3D engines (although there already are plenty of others for his own product available) or game creators instead of pushing the core product forward, then I can only hope that it will turn out to be good for him, but this move really hurts me.

If BlitzMax was available under a FL/OSS license and a MySQL- or JBOSS-like business model, the story would be different, and I wouldn't care if Mark decided to drop the product support altogether: The product would still have an open future.

But as it is now, BRL's focus is not compatible with me, so I can only wish everybody good luck and QAPLA', as the Klingons say. ("Much success.")

I guess that means it works only on GeForce 10800 cards, as the current-gen is 8800/9800.

At the moment a DX8 engine is next-gen for BRL products. ;)

I hope that the syntax of this b3d 2.0 would be more like BB/B3d than like Blitz# ...er... I mean BlitzMax. I know I am probably in the minority here.

Not at all, I like the simplicity of the Blitz3D language, and my first shock when going to BlitzMax was that I needed to specify all variable types.

An intelligent compiler like Blitz3D will choose the right types for you. The only thing which I don't like about Blitz3D is that you have to use that ugly classname\property syntax, where a simple "." would look much better, even the C++ pointer style "->" would look better, but I don't like that either in C++, it should be just "." for everything. I mean if a human can understand what is meant, it's just a question of programming skills to tell it to the compiler.

Personally, I love that feature.

BlitzMax supported handles (assigning types to an int) as an option when it came out. Has this changed?

I love that feature, but hate how it's implemented. Why can't it be like java, c etc.

type varname instead of varname:type

The : makes for ugly looking code, IMHO.

Winni, put your glasses on mate.

Directly from the worklog:

Ok, point for you, Mark. I really hadn't seen it.

Such a small post! How could one... nevermind. :P

Mark clearly said: "The end product will be a multi-platform b3dsdk-like library"

The last update was literally days ago.

The last "packaged" update was 5 months ago (1.28). For those (for whatever reason, and let's not start that debate again) that don't use SVN, that's the most recent update. A further "roll-up" release (1.30) with all the SVN tweaks from the past 5 months wouldn't go amiss.

Winni - hah!

[sound of chalk scraping on blackboard]

1-0 ;-)

a new Blitz3D with the same commandset and docs. in the means of quality would rock the world of indie game dev! It would rock my world at least. I can't wait to hear more information in the future about this.

Thanks Mark, good luck with it!

Blitz3D 2 in my opinion is the only way to go.

For the sake of new ongoing custom an all in one language solution would surely be more profitable and appealing for BRL.

I bet sales of B3DSDK are a fraction of Blitz3D and in similar fashion I bet DarkSDK (what ever it's called) sales are also a fraction of DarkBASIC Pro sales.

I do doubt a plugin 3D engine would generate many ongoing sales beyond the hard core blitzers, but I do feel a whole new language (even updated Max or B3D style) with a capable next gen 3D engine would be much more well received.

The buzz of a next gen Blitz3D would be huge!!! DarkBASIC fanboys crapped their pants when B3D came out, it's time to repeat that :)

Hi Mark (if you're still reading)

Good luck with the new version, we'll all be here to give you a hand with anything when you need it.

I'm amazed how many messages this thread has already!

I am interpreting the worklog as a 3D game library which can be plugged into various languages including BlitzMax, and that possibly the combination of BlitzMax + 3D library could be branded as a single product under the name BLitz3D2. I can see the library also being used from the existing Blitz3D for those preferring that language, but I think BlitzMax is the superior language at this point. A lot of work has gone into it to make it cross-platform and to have better capabilities, it'd be silly to backpeddle and throw that all away.

I am a little disappointed Mark isn't making a game editor type of thing, but that opens the door for me to make one instead ;-D

I don't recall a time when Mark needed help from the community, hes plenty skilled enough to do this on his own. We can certainly provide money though :D

so,

Take Blitz3d

add dx10 shaders etc.
built-in shadows, physics
maybe built-in particle system
throw in b+ GUI
make it also SDK for bmax & others

I'd buy that the day it came out.

Snarkbait, the problem with such a package is that you'd have 3MB executables, as Blitz3D doesn't segregate what it doesn't need from the runtime package.

I can't be the only person who's seriously afraid of Puki.

Not really a problem with today's high speed internet and huge hard drives, really, is it?

Thats just silly. BlitzMax can already do that, but you can select what to compile - so no 3mb exe's.

If this b3d-sdk2 thing gets finished you could probably just link it with BlitzPlus anyways.

The very essence of "Blitz Research", and research takes time.

(I'd still really like to be proven wrong on the whole game-creator thing!)

There are tons of stuff that to me, are creative bottlenecks in Blitz3D...

One of the greatest weakness that I found in Blitz3D are the way we can use Types which I find very limited. Blitz3D isn't really made to play with memory using pointers, which is also a big weakness I find. of course, these could be categorized as advanced user type coding, but when you want to make a game, you get to a point that it is hard to do without, unless you overcomplicate the code. I haven,t looked at how BlitzMax manages this, but the way Types are used in Blitz3D are rather limited.

CSG is another one of the things I wished Blitz3D had. That way you get to make complex models out of CSG operations. So, normally, when you end up with your models, they are supposed to be compatible with, say a physics engine, collision, stencil shadow system, etc... Just like the models you would do in 3DS max! This is what I always wanted. There were some CSG code in the archive, but they are buggy and slow :( But, this is something I wish could be incorporated in the language. Some here might be shocked to learn that Dark Basic Pro incorporates CSG instructions! I never used DBPro but it just stunned me that it included something that, to me sounds so complex.

my whishlistengine .....
include speedimage .bb2
include speedtext.bb2
include physxy.bb2
inlude Guifrom b+.bb2
include Spritecandy.bb2
include Particelcandy.bb2
include Shaderengine.bb2
include internflags.bb2
include shadows.bb2
and moore to come ....

My core concern here is what the target market will be for the 3D engine/library Mark discussed, in regards to the hardware compatibility level of that engine. In the same sentence he uses the terms 'get back to my roots' and 'next-gen 3d engine/library', which leaves me a bit unclear on which market the engine will be targetting.

There are basically two markets to consider. One is the developers who purchase and use the engine. The other is the people who buy and use the end products produced by those developers.

With around eleven thousand registered Blitz product accounts, the main developer market for the engine will be indies and hobbyists (the vast majority will be hobbyists). Those eleven thousand account holders are a good indicator of the target market for the engine, as those people have already bought Blitz products in the past. Most of these people will have mid to low end hardware, bearing in mind that the people who post their system specs on this forum are the vocal minority, and probably tend to have higher end systems than the average user. The majority of these developers will also be the end user of the products they produce, as most people who buy a Blitz product would buy it to putter around with, rather than to produce a released product (irregardless of their original intent when they bought the product).

The main market for the people who buy and use the end products produced by those developers, will be the shareware market. Very few of the developers developing products with Blitz will have the skills or funding to produce a product that would appeal to the cutting edge gamer market. For the vast majority of Blitz developers, trying to compete in that market would be like dropping a guppy into a shark tank. Shareware customers tend to have mid to low end hardware.

Admittedly, next gen technology this year, will be current gen in a few years time. Many of us may have jumped ship to other products by then, however. I think it's important to concentrate on the markets that are willing to buy the product when it becomes available, rather than at some arbitrary point in the distant future when the average users hardware may have caught up with the requirements of the engine. Obviously though, graphics tech has undergone a paradigm shift in the last few years (from the fixed-function pipeline, to shader based rendering) so there's a point of diminishing returns in providing support for hardware below a particular level of compatibility with modern tech.

So basically I think it would be useful for Mark to clearly define (and tell us) what level of hardware compatibility the proposed engine will be targetted at. Mark has talked about producing a cutting edge 3D engine in the past, which I think would be a mistake.

..leave man do his job..im happy to hear that hes moving in direction majority of us want, and thats fine...he will do what he think its best, so why bother..in maintime, go back and do some coding :)

What's the big deal about tiny exes? If you're targeting the power user with their gaming rig (DX10 card and all that), who cares about a tiny sized exe? I'd really like to know if there is a good reason for it... someone fill me in ;)

Sure, but not everone's making a game. If you're writing a simple little application, you want the code to be reasonably compact. There's no real reason why modern compilers shouldn't achieve that (as BlitzMax apparently does).



If you were talking to me there, Naughty Alien, then that sounds like bloody good advice to me.

Hmmm... I guess... but I don't know of many (or any now I think of it) little applications which use dx9/10 and remain small, because to get the most out of that kind of tech you need additional resources that are usually pretty large in size. I'm not saying people shouldn't bring the size of their apps down... it's just that sometimes it's not worth getting worked up about it.

Did you even read that over?

I believe he was referring to things like simple tools or things like decl to blitzmax code converters, bcsgen etc..

Sorry, gotcha. I misunderstood the post where you replied to Snarkbait. Cheers for settin' me straight ;)

What the... I'm away for five minutes and this happens. 8D

Max3D sounds good. I'll even come out of retirement. ;)

All the best with this Mark! If you keep people updated with how things are going and release the odd demo here and there, the userbase will re-inflate to its' proper size. There are a lot of talented people here who are just itching to support this and make it fly. All they need is a solid heading to fly at.

Any programmer who has an ounce of self-respect and takes pride in their work.

My thoughts exactly!

[not my hand, BTW...]

obviously noone at microsoft who worked on coding windows through the years ;)



ive never heard untill the worklog of his "game creator" thing
i thought that was what blitz2d was all about all be it with coding instead of limiting click and play seuck wrappers so going in that direction severly limits the end users to the likes of fpsx10

many many years ago i had the idea of wouldnt it be neat if we had something like seuck for the pc, horizontal, vertical and 8 way scrolling.
i temporarily added it to my to do list of things to eventually do in blitz2d but i decided id be limiting what could be done in the genre buy what i implemented as editor restraints and decided it would just be better to have code that worked on paralax starfields that could scroll in all directions and other 'modlules' that dont seem as limiting (which i never got time to do)

so if dx7 is in for the chop some time this year or two and all our b3d games all suddenly die when users update gfx and dx drivers then im all for whatever comes next to not be so dependant on a fixed dx

Your B3D games will continue to work fine on Vista.

well i was going from all the doomsayers in the make blitz open source saying that gfx drivers dropping dx7 bla doom bla doom bla
i dont even have a hello world app as my work life ballance is tilted too far towards work

Best to ignore those posts as they are misinformed and do not understand how DX works under Vista. GFX drivers no longer supporting DX7 (if it happens (too many XP and older users for that to happen anytime soon especially when the multi-million dollar casual gaming market is literally built on DX7)) has no bearing on Vista. On Vista the DX7 support is provided by Microsoft not the graphics card drivers. On Vista, DX7 API calls are translated at runtime to DX9 API calls, therefore your graphics card drivers only need to support DX 9 & DX 10 for your old DX stuff to work under Vista.

167 posts in one day and the list growing...
I think there is no interest to Blitz3DV2.0 ;)

Good news Mark !

@Moraldi
I thought the Blitz3DV2.0 fanboy thread would be there.

-> http://www.blitzbasic.com/Community/posts.php?topic=78011

As if not all would know that DX7 on Vista is an emulation layer.
But that helps about nothing. Those that play DX7 games do so because their GPU is not exactly Vista Premium ready.
So they are normally not running asta la vista. So it makes no difference what Vista does and does not, if the support is cut from XP drivers its gone and as ATI and NVIDIA unlike Intel still care for XP driver updates, it is a difference to quite some user.
As mentioned: you most likely didn't use Blitz3D that much the past months (especially NVIDIA 100-168 drivers) if you blindly trust in "it will just work" and I hope that your soap bubble just explodes with a band loud enough for you to hear
I give a **** what works and does not work on Vista, I don't use Interface and comming OS prototypes for development or usage. Vista is officially WinME 2.0 and WinME was the worst Windows version ever, at least until MS started to enforce Balloonware Vista 1.0

But as it seems we hopefully will be able to soon work around that and I can't wait to see what the engine exactly means.
I don't assume it to be next-gen (that would be strictly DX10 if NVIDIA and ATI are meant to e supported, DX9 is current and past gen) but a very nice and stable DX9 + Shader support or out of my view even only stable DX9 with AA and Aniso filtering as well as point entity support.

That thread didn't exist until ~140 posts.

Where do you come up with this stuff, since you can't ever provide links to back it up? Still wrong thread for that, it would be better in a myths and hyperbole thread.

dont know if this should be in this or the features thread but as its kinda bout xp and dx7 ill stickit in here cos thats where my other posts were from

as long as b3d2 works on my system like b3d does now im happy
i dont have hardware T&L till i get a new gfx card
but most blitz demo's run fine and dandy

yes some users are going to want to push limits left right and centre but as long as i can hit the rocket button and not get some gfx card error each time then ill be 75% fine (the other 25% being the i want a new gfx card and i need better gfx skills)

From the "features we want in b3d2" thread:

>I bet I won't be the only person sleeping on a triple layer of plastic
>sheeting tonight.

I've always found it fascinating, how every emotion or feeling you seem to have manifests itself in your physical functions, to the point of you actally needing 3 layers of plastic bedsheets.

As for the topic at hand. This is quite a turnaround, and what alot of people have been asking for since forever. If BRL sticks with it, I'm sure they will find it to be a worthwhile pursuit. I didn't think that I was going to buy another BRL product, but if BRL can make B3D/2(all of you old-timers will get that reference) happen, then I am in the target group.

Hopefully the forums will be more active.. :D