ALE Update - Sneaky Peek
Community Forums/Developer Stations/ALE Update - Sneaky Peek
| ||
| EDIT2 : Update is now out. http://www.blitzbasic.com/Community/posts.php?topic=34697 EDIT : Go to this post instead - http://www.blitzbasic.com/Community/posts.php?topic=34670 The demo the shot is taken from is there to download so you can take a closer look for yourself :) Created this in ALE + Blitz in under a day. If you don't know what ALE is yet go to www.dyvision.co.uk  Update1 is free and will be out very, very soon. The base price of ALE might be rising, so if you want to save some money I suggest you get it sooner rather than later :) |
| ||
| Skitchy... Quick question... How do you choose the textures in ALE? I know the texturing is far better than my terrain shader as you're using tiled textures whereas mine just gets one big textures and stretches it over the whole terrain. However, The colouring on that terrain is shockingly bad, it looks totally unconvincing. Are you using algorythms to texture it or is it a manual thing. Also the shadowing doesn't appear to cast shadows... is this correct? Also where you've got the trees there's no shadow below it, I'm aware that the trees were added afterwards in blitz but is there anyway to say, "I've got a tree at x,z can I have a blob shadow here please?" within ale or to affect and ale terrain. Don't get me wrong, Ale looks like a fantastic app. Just got a few questions. |
| ||
| There *are* shadows below the trees, I just chose a bad angle (the sun is behind the camera. My bad :) Follow the guys nose to the first tree you meet and you can see the shadow. Masked textures are taken into account, so you get shadows in the correct shape. The trees were *not* added afterwards. That's one of the new features in the update - adding 'props' to the landscape (note the lighthouse too). All the trees were added using a special function that takes about 5 clicks :) You also have full control over each individual 'prop'. You choose textures in ALE by clicking the 'Replace' button next to the appropriate texture in the toolbar. You can also choose the color 'tint' you want by clicking the color box next to the Replace button. And the slider below each texture sets the UV scale. The terrain is all 'painted' by hand in realtime (that's what ALE is all about). You can import/export heightmaps if you want to though. |
| ||
| Dare I suggest that people like myself who are not so good at art would want a nice quick... "Paint the terrain please" button, then after it's painted be able to edit it afterwards? Even if it's just something as simple as terrain shader so you at least get the low/med/high and steep slopes painted with 4 textures then you can go wild and add your roads and so on? Just thoughts... |
| ||
| Just use a big brush :) Honestly, it really doesn't take long. I spent more time finding textures than I did painting the landscape! |
| ||
| I'm not talking it taking a long time or a short time, I'm talking about putting the right texture in the right place, personally I wouldn't care if it took 10 minutes to work out where to put stuff if it came up with a good realistic representation. Check out this:  I know the texturing isn't as precise or as much detail, but because it's textured using an alogorythm the steep textures, low/high etc all look correct. Personally I think this terrain looks better than yours, however, I know for a fact that your terrain is technically light years better... So my point (and there is one) is that if you add this kind of functionality to ALE the terrains will look absolutley stunning. The reason why I mention this is I'd certainly want to buy ALE if it had this functionality, I'm guessing I'm not alone. If you want some help then I'd be happy to lend a hand as ALE looks superb! I just think it needs this aswell. |
| ||
| Hi there, Woot i can't stay for see this Update1 :-) Kiss. |
| ||
| I agree with Rob, some automation based on height and/or steepness would go a long way. [EDIT: that is the craziest sky sphere I've ever seen! :) ] |
| ||
| can you get some translucent water in the shot ? |
| ||
| Sorry but if you even higher the price then ALE is off reality with it's pricing ... although it's result is nice, it isn't really hard to create something like that. |
| ||
| :) |
| ||
| Skitchy should slash the price in half. And then he'd make more than twice as much money due to increased sales and there's no overhead in selling 1s and 0s now is there? Just a friendly suggestion. |
| ||
| As long as everyone has "earning money" on position 1 and "program something allround usefull" on position 2 the tool is useless anyway ... at the moment the terrains are nice but not more still core stuff missing but I've suggested them twice now, I'll write my own editor for that use, is no problem, I've been playing with it in realtime for weeks now |
| ||
| Funny how anybody who's written a half-arsed 'dropper' seems to pop up on an ALE thread trying to put people off while subtly trying to pimp their own stuff - don't you think? And half a game (£20) ISN'T a lot of money. Oh, and I'm selling plenty thanks - probably something to do with the program being good. The world is crazy like that. ;) |
| ||
| or with the fact that there is no other program so far that does it or that the old ( just to mention ) texture platting technology seems to be to complicated to learn. As said, there is stuff missing to make it allround usefull. So far it is only usable for simple mixed outside terrains but not for a real outside scenario. If ALE would have been doing any real complicated stuff or programing efficience sensitive stuff like Gile[s], then the price would be ok ... but 40$ for a 20% usefull terrain editor that every 2nd person can write on his own? perhaps it will be worth that much somewhen, but it isn't for the moment, thats sure |
| ||
| Hahahahaha - like Homespun is a work of genius - LOL. If so many people can write it then why am I the first? You're nervous, and nervous people lash out. Your unfinished product has been blown away by something vastly superior and you're crying about it - simple. Go ahead - if you think you can write ALE2 then give it a try. I'll be here waiting, but I'm not holding my breath. Now stop trolling my threads and go code your little dropper. |
| ||
| [sarcasm] It is extremely hard to write something that construct 6 copies of a 129x129 vertex mesh, changes the vertex alphas for a smooth texture transition blending and does an optimation before export.[/sarcasm] just for simplicity you would have added a 2nd mesh on export with a combined colormap which people can use for TerraEd, Gile[s] etc because your terrain does not work caused by its very specific setup. If your stuff would have been thought through far enough, you would have thought about that and planned the system in a way so people can use it with existing and more and more becoming standard apps. ( instead of saying that it is all others apps fault ) Same for the suggestion/request i made about 3 times now: you forgot to add a possibility for a non-blended layer which is something that might just be needed for stuff like roads etc ( for stone paths it might not be needed but for more it is ). Thats another thing you never seemed to think about. ALE is really nice but 1. overpriced 2. it uses a outdated technic ( vertex alpha blending instead of texture alpha splatting ) and creates meshes that only work under very specific setup 3. You don't seem like you did ALE for a specific project, I don't see another reason why it is not compatible to most other apps and is lacking core functionality out of real project usement sight |
| ||
| @Dreamora It DOES use terrain splatting - that's what the optimization stage is for. If you had 2 braincells to rub together you'd realize that. You've obviously given ALE a [sarcasm] REALLY [/sarcasm] good testing, beacuase you're WAAY off with the vertex count there too. ALE is compatible with Gile[s], TerraEd, and anything that can load a .b3d. The setup is a standard .b3d file - oh sorry, that too specific for you? If you want to see it working in a game situation I've posted a demo FPS in another part of the forum. If you can figure out how to work your mouse, go find it. If you think its overpriced then just don't buy it - nobody is forcing you. And if you're as clever as you make out I'm sure you can mould a road mesh to a landscape (its not exactly difficult). Or go and write your own ALE - go on - please - I'd LOVE to see the result. Make sure you post it on this forum so I can give you an unbiased, balanced opinion as you have given me. Now stop trolling. F*ck, why do people treat me like this on this forum? Did I deeply insult everybody by releasing my program? Should I have just kept it to myself? |
| ||
Dreamora - stop being a .. .. and let Skitchy concentrate on his excellent application. The only person trolling here is Dreamora, everyone else is giving constructive criticism. Rob Farley has even offered to help out. :) |
| ||
| Funny how anybody who's written a half-arsed 'dropper' seems to pop up on an ALE thread trying to put people off while subtly trying to pimp their own stuff - don't you think? I'm not trying to pimp my own work, simply trying to improve yours, I've already said your system if far in advance of my own. I take massive offence at your comments. It the first time I've heard droplet described as a half-arse 'dropper' but thanks for the review. |
| ||
| The 'half-arsed dropper' wasn't aimed purely at you. Ever since I released it there's been a troll-fest by various devs around here and I'm sick of it. I believe that last comment is a banning offence. |
| ||
| (Apologies for the last comment, I was very angry at your comments, edited) What is your problem Skitchy? I have no grief with you and I thought I was being constructive. However, if you don't want to accept help or critism that fine. [edit] I think £20 for ale is reaonsable. I think Ale looks like a top product, I think Ale could be better with a few additional features. Which I've pretty much said in my previous posts. [edit2] Even after all this I am still seriously considering buying ALE. [edit3] Probably from a pub though! :) |
| ||
| I agree with beaks. Dreamora=bellend |
| ||
| Hahahahaha - like Homespun is a work of genius - LOL. Why say something like that? HomeSpun is something anybody could write but it works and saves some people the bother of having to write it themselves. I can't believe you'd say something like that? I guess you speak for yourself and it says a lot about you. I would've probably bought a copy of ALE like I said earlier I was going to after you added what you call "dropper" functions, but ...uh, I don't think so. |
| ||
| oh my goodness... --Mike |
| ||
| Well, each of these systems I COULD write myself, like any decent Blitz developer should be able to do. But then there is the choice if you WANT to code it yourself or save time by buying it. For me, the fun comes with coding stuff so I decided to code all my tools and systems myself. Other people just want to go ahead and start on a game, which is why most of us bought Blitz for in the first place. Don't be a jerk and state the obvious by saying you can code it yourself and therefore another person's effort mean nothing at all because, frankly, that is not true. A lot of code which makes up these tools appear in the Code Archives and are a great help for others do decide how to approach a certain problem... and that is invaluable. |
| ||
| Trolls be gone !! :) I just bought ALE and couldnt be happier :) The TerraEd demo running at 8fps and crashing my system any time I tried to make changes clinched the deal :) As for you that write your own tools -- dont you value your time ?? (Same goes for those that claim they could write it, but somehow didnt beat Skitchy to market !!) I contract out programming work at $100 an hour, so lets see, $40 USD I just spent works out to about 24 minutes of my time -- dont think I could code it up quite that fast !! what is your time worth ?? Skitchy -- Remember my request one of these days !! being able to draw an open or closed curve (perhaps in an overhead view of the entire terrain) and have a specified width along that curve smoothed/leveled to be a road/path :) Would be even sweeter if it could optionally generate a mesh prop with a specified texture :) (I dont ask for much, do I ??) I've got some ideas how the UI could be worked out if you want to hear them :) |
| ||
| A lot of code which makes up these tools appear in the Code Archives and are a great help for others do decide how to approach a certain problem... and that is invaluable True... |
| ||
| As for you that write your own tools -- dont you value your time ?? (Same goes for those that claim they could write it, but somehow didnt beat Skitchy to market !!) I contract out programming work at $100 an hour, so lets see, $40 USD I just spent works out to about 24 minutes of my time -- dont think I could code it up quite that fast !! what is your time worth ?? You can't compare hobby as a hobby and contract work. And tools made for MY games are not meant for the market. Think before you post. |
| ||
| ohhh myyy goooodnesssss!!!! and this used to be such a friendly place to visit... like a community of sorts... only with friendly people... and pretty flowers, and stuff... --Mike |
| ||
| oh Red, are we really that bad? And have you read your own posts lately? Lots of negative vibes there. I'm not going to react to your reply here as it is off-topic already. Mail me if you want to talk 'positiveness'. |
| ||
| I always think before I post, about Return On Investment :) Wether that return is in "money" or "fun", I dont waste my time unless there is something to be gained. And reinventing the wheel is always a waste of time. Since I do contract programming for a living, I'm balancing time I could spend making $100/hr vs time having fun writing games. So yes, I value the time spent on my hobby the same as I do my time making money. Skitchy -- Thanks so much for saving me time and money either creating the tool my self or using less capable substitutes :) |
| ||
| Interesting problem -- wonder what I did wrong :) I used the default ALE, lightmapped it, exported it and the props file to a new directory with textures. Got the textures ok, didnt get any of the b3d meshes that the textures applied to, and the props file had no paths in it. Did I mess up or is props export incomplete ?? Yet another suggestion -- export option to auto-name all entities using the filename sans extension + 3 digit number ?? i.e. tree001 tree002 etc. Gonna be easier at this rate to make an integrated export that just lets you checkbox the things you want, rather than seperate exports for mesh and props and lightmap. btw -- if I dont load props, everything else loads fine, and looks good in my walkaround program :) Nice and fast too :) |
| ||
| ZombieWoof I feel sorry for you ... counting fun and hobby the same as work must be a really horrible life nice done skitchy |
| ||
| The default level is actually slightly different from a standard ALE file. If you create a level from scratch, loading the textures and models from your own drive then the 'Copy media' function should work fine :) You can then save this as your 'default' level if you want to. |
| ||
| @Dreamora -- Just because I use different standards and methods to decide how my time is best spent does not make my life horrible -- I'm quite happy with it -- I make enough money to pay the bills and support my family on 20-30 hours a week, I work at home, and I spend lotsa time doing fun stuff (like programming games) -- doesnt sound horrible to me at all :) @Skitchy -- thats weird -- but ok !! I woulda thought it wouldnt matter where the mesh/textures came from, since it did copy the textures for the props. Just not the meshes. Watch out !! its suggestion time again !! stick something in one of the menus to remove all props from the current terrain !! (do let me know when you get sick and tired of the 2-3 suggestions a day I been throwing at you since I got a full copy of ALE :) ) |
| ||
| I *never* tire of suggestions :) Just don't expect them ALL to make it into the program ;) 'Remove all props' is a good one, and easy to implement so I'll add that soon. Thinking about it, I could also add 'remove by name'. Say all your trees had the word "tree" in their name ("tree01", "palmtree", "dead tree") - you just type "tree" into the appropraite textbox and it seeks and destroys them all. Sound good? BTW, the default.ale file needs to be different because standard ones store the absolute path to textures/meshes on the users machine. The default one needs to work on ANY PC, so just stores the filename. Exported meshes obviously use the stripped filename as well, because they need to work on multiple machines (ie. the gamers PC). |
| ||
| Ok -- todays suggestion -- kb shortcut for delete prop !! Mousing around all over to delete is tiresome :) (probably should do something for the yes/no box too) Hope some of the ones I dropped on you in email make it :) |
| ||
| Odd behavior -- when moving a prop, the yaw and roll jump back and forth from min to max, and the original setting is lost when you stop moving. Should be an easy fix :) |
| ||
| I've never seen that one here, but I'm pretty sure I know what's causing it :) Will be fixed. |
| ||
| might have something to do with it being a huge 40,000 poly prop :) Gotta have a long talk with my artists about LOW POLY !! Of all the other suggestions I've given you, I think the one that would do me the most good right now is the ability to "push" a prop into the terrain. So I'm suggesting that one again !! I've figured out how to loft race tracks but adjusting the terrain around the track is very time consuming :) Would also be nice for automatically leveling the terrain under a building :) |
| ||
| Procedure for levelling the ground under a building : - Set brush size to approx the same size as the buildings base (or just use a smaller brush and 'paint') - Use 'Flatten' to level the ground - Click the little "G" button next to the "Alt" slider on the right hand side of the toolbar. This auto-calculates the altitude for that prop (taking scale into account!) so that the prop sits directly on the landscape ie. not hovering or buried. Hope that helps. By 'push' a prop into the landscape, I assume you mean deforming the mesh so that it follows the same contours of the ground (as opposed to actually 'burying' a prop). I'm trying to decide the best way to do this so that it works for everybody, but you *can* do it right now. Assuming your racetrack is built in the same way as a terrain (ie. a tri can never point 'down'), all you need to do is export a heightmap, then apply it to the racetrack mesh as some sort of height deformer in whatever modeller you are using. :) |
| ||
| Other way around !! I meant deforming the terrain to accomodate the prop mesh, using the brush size (or 1/2 brush size) to figure how many verts on either side of the prop to slope the terrain to meet the prop. The prop in question is almost as large as the terrain, so hand editing is very time consuming. like this: Cast a ray vertical up and down from each terrain vert to test intersect with the prop, if intersect mark that vert, and all 8 adjacent verts as being under the prop. Set all these verts to altitude of prop. (optional: if the prop has sections with varying altitude, handle the offset from base altitude of the prop to the altitude of the actual intersection and bring the terrain right up under the raised portion of the prop) Scan all terrain verts, if they are within 1/2 brush of a vert under the prop, figure the distance to the nearest covered vert and set altitude according to (average of brush edge verts not under the prop) averaged with prop altitude (or something like that to make a slope up/down that converges with the terrain outside the brush, you may have some better ideas how to make that happen) I could probably do it in the modeling software, but I'm spending way to much time in the modeler, instead of in the editor writing code :) It's gotten to the point where I know more about high end 3D modeling software than my artists do, but I still suck at the creative stuff -- they work in Poser, not 3DS/Maya/Truespace :) |
| ||
| Ahh, I know what you mean now :) Again though, you could use heightmaps but in the opposite way to how I described. Export a heightmap of your race-track mesh from you modeller and then import it in ALE as a starting point. Then you just need to do the fine details/add hills+valleys where there is no track. Your idea is good though, and I'll give it some thought. |
| ||
| Doing the fine deatils would be a lot easier if (watch out -- another suggestion !!) you could lock the terrain verts under a prop :) Perhaps a new "brush" type (ala flatten/raise/lower) that lets you lock any verts if you paint over them -- turn them a different color too :) |
| ||
| "The TerraEd demo running at 8fps and crashing my system any time" I'm sorry but i have sell many terraed copy and you are the one to tell me that ??? it's strange ? Terraed was coded, on a P4 1.5 Ghz + Geforce 2MX and run well ? it's strange. |
| ||
| I never have any problem with TerraED at 8fps :) change your comp :) Kiss. |
| ||
| TerraED and ALE working well there too. Regards. |
| ||
| The problem is the ATI Radeon Mobility M6 on my laptop -- its got incomplete 3D acceleration and the driver apparently reports incorrect capabilities to the D3D HAL -- I have problems with lots of 3D apps :) Its still my primary workstation though, my benchmark for "minimum requirements" for anything I write. The monster machine (listed #2 in my sig) is the file server and render farm for movie clips and cut scenes. Nothing against TerraEd -- I still may buy a copy later on down the line -- it just wont run on my laptop -- expects some capability of my vid chip that it just doesnt have. |
| ||
| Oh ok i see :) good luck ZombieWoof ! Regards. |
| ||
| gonna need it -- my artists are balking at learning the new stuff they are gonna need to do 3D game art -- they are coming from a poser/flash environment -- and I'm a programmer NOT AN ARTIST !! heheheh :) which explains why I'm bugging Skitchy (and others with useful 3D apps) for all these additions so I dont gotta learn 3D modeling :) |
   |