new VertexBuffer() Xna3.1 -> Xna4.0

My latest errors seem to revolve a lot around vertex buffer api changes. There is  nice blog post here by Shawn Hargreaves.
Post about Vertex Buffer Changes

simply put though you remove your VertexDeclaration lines of code and just put 'typeof(RoundLineVertex)' the declaration in the second parameter of the VertexBuffer call, like these following two lines of code.
vb = new VertexBuffer(device, typeof(RoundLineVertex), numVertices * bytesPerVertex, BufferUsage.None);

// vdecl = new VertexDeclaration(device, RoundLineVertex.VertexElements);
 
After going through that it was very easy to change the VertexBuffer setups... Only now i have IndexBuffer to sort out. I did not write this original code and so have no real working knowlegde of the finer points required to make this stuff work. However I hope it wil be fairly straight forward once i under what has changed in the API. I understand what the buffer is for, only not whats changed at the moment.

Laters

Blooming fixed some more Xna3.1 -> Xna4.0

So i had 28 erros, ALL of them pertaining to the Bloom component. Interesting. I know by experience that this is not the last of the errors. They are just being masked by the current issues. So i was not surprised when i fixed the Bloom to see a new selection of errors, 48 in total.

So fixing the Bloom was easy. I just compared the New 4.0 sample to what i had and copied over the changes. I would have just copied verbatim, but i knew i made several small changes to the Bloom and i did not want to do that again. This process only took about 5 minutes anyway.

So now onto my new Errors which include
does not conatin a definition for 'MinimumVertexShaderProfile'
and
VertexElementMethod does not exist

These are all to do with Roundline which i am actually not using at the moment in this game. I may remove it for now. I will bring it back when i want to use it (gonna be used for the Menu button descriptions I hope!)

So ever forward we go.

Now tomorrow night is my works party, i am sure i will not be posting then :)

G'night

Xna3.1 -> Xna 4.0 Solving missing types

So i had 158 errors and the most obvious one was the mising definitions for certain types like..
Vector2
Vector3
Vector4
Rectangle

To solve this issue i created a brand spanking new Solution and opened it up. As the problem appeared to be mising references i started by opening up the referecnes for the new project, then comparing them with the ones in my broken project. Immediatly i saw i was missing a couple of very important reference files.
Microsoft.Xna.Framework
System.Net
Obviously without these there is no way it can compile. You can just right click the References line and say 'Add Reference' then simply browse for these two to add them in.

Onto the next issues later..
Oh, i have about 28 errors left, Yay!

My Motivation is Lacking

Or put another way..
I do not yet have enough motivation to trawl through the myriad upgrading problems presented to me by the upgrade of Xna 3.1 to Xna 4.0 game Studio.

Still i feel every day more and more resolve, only not enough yet to actually boot up studio and stare at the current issues. Last night was close but lost out to spending time with my wife.

So the last week or so has still been busy for me. Playing lots of Cataclysm which i have been enjoying rather more than i expected. I have also been Kinect playing with Dance Central a lot. Boy is that my kinda game. In fact for me it is my GOTY. I even parted with 6 stinking dollars to get 2 downloadable dances, bah! I hate you Xbox (well not really). al i need now is to get a few more friends gamertags that actually play this game. Comparing my score to one friend is ok, but not enough darn it! I need to beat or be beaten by more people. It is also great cos my wife, son, and daughter also like to play this game. Truly awesome. I hope one day it will be able to actually play two at the same time (not like the take ya turns as it is now).

Maybe tonight will be the nigth i power up Dev Studio. I feel i have more resolve and also a few more tricky ideas of simple thing to try that i believe will solve my current issues. then onto new ones.

Laters

So i removed the Avatar code (Xna 3.1->4.0)

After all the issues last night i have decided to remove all Avatar code for now. The chnages look just too all encompassing.and i need to get something going, if only for my sanity. Though sanity aside, it is always best to divide and conquer a problem if at all possible, and i am of the beleif that all problems can be broken down in some way.

So what happened after removal of the Avatar code? Well i ended up with 158 Errors, woohoo! Now there is something i can dig into, only i wonder how many more are hiding under these, as is usually the case.

So my next Error is... (unexpected)
"Error 1 The type or namespace name 'Vector2' could not be found (are you missing a using directive or an assembly reference?)"
I also get similar errors for Vector4 and Color (now i knew Color had changed but the Base Vector type declarations?). I guess it's time to see what i should be presenting these as.


After about 25 minutes searching, i have not found an answer. I have a couple of the Samples from the App hub and they seem to work perfectly. So i have no clue at the moment what is wrong with some of my types. I also see the same issue with 'Rectangle', i am sure one cure will fit most of these.
 
Off to work now. But if anyone knows what is wrong or even a lead, i would love to hear it.

3.1 -> 4.0 Xna First Errors

My First Errors... Most likely what you will also see.

So what should i attack first? The component link issues or the Pixelshader. I guess i should go with the component and unleash a real torrent of errors :) It is more fun that way. Fixing the Pixelshader will only solve one problem.

• So i simply right clicked the references in the project and deleted the offending system.xinq.xml.
• Microsoft.Xna.Framework could not be found.  So i did the same with that. Not sure what repurcussions this might have later.

So onto my Pixelshader issue. While looking around the web for help on this i found this most excellent Blog where someone has done what i am doing now. Check it out here. So the change was quite simple. I just had to change the pixelshader name (yeah the bloom was using PixelShader to PixelSdr and i also saw that it had updated the 1_1 to 2_0 shader version. So onto my next issue.

My next issue that appears is becaue i am using Avatars. And as such on the PC it draw a set of polys to show the bones of the Avatar. Great for making sure you can prototype you game on the Pc, but bad cos now i need to fix whatever is wrong. The offending error is....
'XNA Framework HiDef profile supports a maximum of 16 simultaneous vertex elements'. This has caused me to figure out the .FBX animation files for my avatars are not compatible with Xna 4.0. So here is a link to the new ones New Avatar Anims. Yet still i am having issues with the Faint.fbx. the very same error as before. Not sure what to do, maybe i will attack it tomorrow night, so far i am losing this battle. I could just remove Avatars for now and re-introduce them later, but i still need them. I will also have a look at the new Avatar sample program for Xna 4.0 from here. Oddly enough the Avatar sample does not have errors in the same Faint.fbx file, though it does have a lot of warnings as expected. not sure what the solution is at present.

That is all i can do for tonight. I know i am not the most expert at this kind of upgrade, but i am also sure not everyone is great at this.

Time commited so far... 1.5 hours

G'night

Getting Xna 4.0

Should be simple enough Dotcha think ?

Just go to the downloads section of the App Hub and there ya go? Well not so simple.

Here is the link and what it says on App hub. Some more links that seem to confuse more than solve my issues. Really, why would you not at this point just give 2 clickable options? One for Phone 4.0 and one for windows/Xblig.
What ya get on App Hub

This is the link everyone will want that just wants the Game Studio 4.0 for developing for Xblig and Windows.
Xna Game Studio 4.0 download link

This next link is where you can find all the Visual studio express downloads. Much easier than trying to find an actual link on the app Hub.
Visual Studio Express Download Link

Oh, i installed the Visual Studio first (i found that instruction somewhere and felt it made sense.


Next post should be me attempting to solve some of the first problems that we encounter in the upgrade. I intend to Blog the whole procedure to make it faster and easier for the next person.

All Games Are Crap part 2

Not content with writing about games being crap just once. I home come back to do it again.

So if it isn't the good or bad in games that allow them to do well, what could it be. Why is it that some games do so well compared to others in the same or similar fields. Well to start with i want to say, advertising. Yup, as simple sounding as that is, it is not simple. In fact there are many aspects to advertising a game. From the posters and artwork you see, to screenshots and carefully leaked ideals of gameplay that the new game brings. some may say that advertising (and marketing) do not play a huge role in the selling of a game. I have to refute that idea right away. Why would a large publisher spend 2 million dollars advertising and promoting a game if it did not need to?

In the end a game must be seen by the potential buyer to actually get them to buy it.

Of course advertising a game that has a big 30 million dollar budget is a nescessity, as seen by the huge budgets i just mentioned. What would the sales be like of Darksiders for example if they had not spent so much on blanket advertising and marketing. Just getting that game to E3 must have cost a fortune, and in theory that is just an industry event (yeah right!).

Of course we have a microcosm in the Xblig area. Though all the devs know that visibility is still the key component in selling games, whether they be bad or not is less relevant. Having 10K downloads and selling just 5% is much better than 2K downloads and selling 20%. Of course i do believ something else...

Quality of presentation and gameplay, including music makes a better conversion rate.

This also includes crap games. It is possible to have a simple game that will do well because of it's presentation. I would offer up a particular Avatar game as evidence, but they have had plenty of press recently. The point is that it looks nice enough, the gameplay is ridiculously simple. Yet their conversion rate is high, and why not.

So what can Xbligs do to get the word out about their game? That is a good question and i have been working very very hard this year to try and get word out. I have failed, but i try. One recent group has apparently resorted to publicising their new application by ridiculing the Xblig community through Kotaku. this appears to be a brilliant move. Unfortuneatly i am not very good at that good cop bad cop and being the bad guy thing.

So far the only answer i have to the visibility is to have an Avatar in the title and game. so i will continue to play with them until i have a good game to put them in. I cannot just put them in for the sake of it, like a number of current Avatar games have.

Well thats all from me, Get a huge budget and maybe it will pay off, i have tried spending, but cannot seem to find the Xblig buyers.

Good luck to everyone else

G'night

Some web sites make me feel good

Here is a web site that decided to offer  something that everyone with an Xbox should be interested in.

I have to say i am very proud to be on that list.

LIST of Xblig Worth Checking out.

Apart from that i have to also mention that my job is currently kicking my Butt. I am working extremely hard trying to reverse engineer some combo binary data packets and then grab the vert data that is in them. I am even having trouble finding vert dtaa at all, never mind extracting it correctly. Now these are binaries created for a PS2 game, but geez, i usually handle this stuff no problem. Instead i feel weak and annoyed.

I certainly hope tomorrow goes better, as two weeks working on deciphering this is getting real old, fast!

G'night.

What Next ?

Is anything next? at least for my lovely Xblig?

At the moment i think not. With only 2000 downloads it seems very few are interested in what i do. I will however do a post mortem as soon as i can. detailing decisions i made for the game that have ultimately stunted its spread.

g'night

Competition Event #2 Winner

Here is what i saw at 9pm pst tonight on my xbox.

A very definitive win for sure. Placed directly below my own new score and my artists score. I am very pleased to see someone win who has improved on last weeks score. This makes me happy people are improving at the game.

Well after all my internet and virus problems this weekend i have to call it a night. I look forward to sending off a T-shirt to Germany no less.

Don't forget that next weeks competition is in a wekks time.

Good luck and please pass on the competition info to anyone you know with an Xbox360.

G'night.

A Shooter Video is done.

Yep it is true. After too many hours i have the new promotional video for A Shooter. Check it out below.



I don't think it came out too badly. My only real disappointment is that i am only showing off a fraction of the enemy types that are in the game.

I also noticed that Youtube have not used the default picture i told it too, bah! Now it is a generic frame from the video, that is annoying...
Laters

A Shooter enters Playtest

Should i Doom my latest playtest version by saying that this will be it's last time in there before Peer review.. Yeah i guess i just did.

There is a very large discussion going on about high score sharing on Xblig at the moment. I have several comments in there myself. There are a few groups of Developers or opinionators on every side at the moment, and no clear winners so far. Although i feel that some posters feel they have won or are winning something. Mostly i just think we are just headed for more rules, more restrictions and a lot more time in development and also review.

So remember, it's all for the children. ;)

I am done with this topic for now, though i expect to ruffle some feathers with A Shooter as it does indeed have High Score Sharing. and not in a way that some of these Peers feel is acceptable. I will take my chances and see what goes down. I will not break the rules for my game, however i will argue what those rules should be.

G'night

The Promo Video - is it important?

There seems to be two theories on this that i can ascertain from reading blogs and forum posts, and also talking with other knowledgable people.

1. It makes very little difference
2. It will help sell the game

These two are not opposites and can certainly both be true. In fact I personally believe them both to be true. Though there is one aspect to making a video of your own game that goes beyond money. It is 'Worth'. If you believe in your product, then you should have no problem being able to come up with a short Promo video for it. If you do have a problem, then maybe something is missing or just not quite finished yet.

I make videos for all my products. In fact i rather enjoy the challenge of putting al my work into about 1 minute and thrity seconds of video. Adding titles and overlays and transitions as i think will help make it more appealing and palatable.

I now have a basic script for the A Shooter video and have already recorded all i need (or i believe i do). So tomorrow i get to start on editing the video together. Chopping and hacking and splicing a small video together from about 20 minutes of footage. It is a time consuming thing to do, but i am usually proud of what i end up with. as i was especially so with Pellmell.

1. If it makes little difference because noone sees it, then it is what it is. I will put it out there and shout about it the best i can. Publicity for an Xblig product is just hard to find.
2. I think the video if done well will certainly help sales. It will mostly help downloads. Though again, i will need people to see it.

G'night.

High Score Sharing - Done

Yes, you read that right. It is now done. Working really well and the hiccups are very minimal. I cannot say entirely removed, i think because of network session underlying code can do that. However the stuff on top appears to be clean and working.

I found my bug. I needed to make sure i waited a second or two before terminating the session from the Host. For some reason this caused the ElapsedTime field of Gametime to get really really small. As there is a timer on the sharing code to only do stuff every so many seconds, then it basically stalled me out. Once i sent in a fixed time delta (1/60th) it updated fine as well. so with both these things fixed it all works very nicely.

Testing it has taken some time. I am methodical and commited to doing that part properly, so i have left it running over several hours to see what it would do. And it works great.

The last thing i added was a Previous Hosts list. This is a class that will store anyone you connected to (the host). Then there is a timer on them. This effectively causes a timeout from matching up with the again. I might set this at 4-5 minutes, i am not sure. I could see this being game dependant though. for examply 10 minutes playing my game is quite a lot, but 5 minutes is low. So i could set my previous Host expired timer to soemwhere between those figures.

I intend to share this code once i have finished tidying it up. there are plenty of comments already, but some of the names are a bit wrong. Once it is done i shall make it open source. Oh.. and it is about 600 lines of code (including comments and lots of debug messages.

G'night
Tonight i go to bed a much happier man. Next up is encrypting my save file.

Networking errors can suck time

And that is the reason for my lack of posting this last day or so.

I have borrowed an xbox off a friend of mine to work on the shared highscore stuff. I needed to make it garbage collection free so i could run it while playing my game. The previous version was so bad timing suck wise that my game would judder badly for a couple of seconds, totaly crapola. The new version works seemlessly i am proud to say.. and i got that running very quickly. However i have a problem that is odd in nature.

I would go into details but i am so tired now. So many hours watching this thing die is very demoralizing, when i was this close a couple of days ago. Such a waste of a weekend, but i truly believe that a shooter should always have a real meaningful high score system, and as MS don't give me one, i will have to get this networking code fully functional.

so over this weekend i must have spent more than 12 hours working on this one bug.

Bug:
A Joining machine find s a host and receives the scores that are sent. Then he reciprocates and sends his high scores. all good so far... Then....
Occasionally the joining machine will end his session, and hang there. The update call i have just doesn't seem to happen. I get no debug prints or anything. I believe the debug prints stop because all threads have locked. If i break into the code (anywhere) and immediately set it going again, the threads appear to be reset and it all goes along on it's merry way again. Till the next time.

I cannot find any useful (at this time) info on the web about this. The network session is just sort of getting stuck, and as a high priority thread, well it kills all else.

More tomorrow i suspect. so far my latest attempt is up to 5 mins, but i think it is just messing with me.

G'night

Xblig - A Shooter and more

It has been a very busy bunch of days for me this week. I have been very productive though and hopefully soon Xblig playtest will see some of the work i have been doing.

I will start with A Shooter...
I have been rewriting the highscore sharing class, even though it works in my game. The reason for this is simply because i cannot score share while actually playing my levels. Due to the massive garbage collection issues it stutters my side scroller so badly for a second or so, you will mostly die. That would be unfair and go against my friendly none cheating game policy. So i have rewritten the component to make it so there is no garbage collection (ok, there is a teeny tiny bit). Now i knwo i could have updated to the version2 component, but to be honest. When i lokoed at it, it was huge. I don't write huge anything. I write simple small amounts of code. So i have it mostly working and tonight when i have obtained my friends Xbox, i will be testing the score sharing once more. Efficiency is the name of the game here. A couple of frames hiccup ill be ok, but any more than that and i will have to work on it some more.

I think that after i have gotten this all working i will spend a bit of time tidying up the names of some of the variables and then put it here on the blog for people to use or reference if they want. The reason the names of variabels are a bit odd in places is because they were there while i was cutting the code base down. After it works will be the time to name verything to be more explanatery.

Tomorrow i will post about A Dreamland Chronicles Game update and the idea of Episodic content.

Laters

Intro Video On the Xbox Issue

I got home tonight and spent some good times trying hard to Code 4 Kris Steeles new game Hypership Out of Control. He just got it through peer reveiw and Bam, one of the reveiwers finds a code 4. This is serious business as it in theory just went onto the new releases list I have no idea how that will lay out, i sure hope he gets it on the list when the game is fixed. In comparison to Kris's problem mine are basic.

So i wanted to get my new intro video onto the Xbox, to see it in all it's glory... Only i had to wait while Hypership was being tested (i think my game was crashing others, so i left it up). The waiting mean't i got some other coding jobs done, so all good there.

However when i got the intro onto the Xbox it was al choppy. ARGH! so i went straight to the video editor and worked on smaller less Mbps Wmvs. In fact i have ended up spending more than an hour and a half, thanks mostly to the export times of Video. sometimes they would generate corrupted stuff. Others they worked great on my PC. Then when i put them, in the game and ran from the solution, they looked choppy and crap. It took me quite some time to realise that when streaming from your PC, it is crap. Only when i decided to work on a trial mode feature and launched my game fram the Dash did i see that the video was in fact alright. which means i have wasted sooo much time tonight :(

So the lesson is,
Don't believe all ya see.
Video running through the debugger will almost certainly look awful.

g'night

Texture Swapping FTW

Working in Home on the PS3 has been quite interesting because of it's restrictions as much as anything else.

Ya see, to get the best from a next gen machine you really have to push some major buttons. You have to work quite hard to make the machine actually sweat. Managed languages put an amazing strain on the systems that they exist on. Even the most efficient ones do.

Then we have working within this managed language, it is like you have a machine but it has had restrictions placed on it, making it less of what it was. This means it is easier to work with, but also easier to reach it's limits. It works out like this on both the PS3 through Home and the Xbox through Xna. Whenever there is an update to the SDKs i get excited. What kind of new goodies have they brought me? Avatar support for Xna?  Rendering control for Home?

So today i tried one of the new features expanded in the last Home HDK, the power to Texture swap. This is an amazing thing and opens up so many possibilities. It also makes art pipelines much more streamlined for replacement of stuff, such as the digits on a digital clock. Now with texture swapping it is real easy. Less work for the programmer and artist both.

It may seem like small step for some, but to me it is very exciting. What amazing things am i goign to do with an artist and this new found power... Only time will tell,

Laters

ps. My family is now back with me, Woohoo! I am soo excited.

ARGH! or is that AARGH?

So i have been moaning today. What i have been moaning about today might be a trifling thing to a lot of people, but to me it feels quite important. I haven't put any pictures lately into my blog posts, as i have been away etc. Well today you get to see someone elses artwork (i could call it that).

I like Xblig, i like some of the community, and i certainly some of the games that are there. It is an amazing place for so much good stuff. However it has a bad reputation, and one i gues it might truly deserve. It really is hard to find the good stuff from the crap, at times. Sometimes it is real easy.

So i finally got some peer reveiwing done today. I had a rather good time and it only took me 3 hours for 3 games, and one of them was very simple. So after i had finished my third game i went looking for a fourth,and it was while i was looking that i noticed somethings horrible.

These pictures were in the list of games to be Peer reveiwed... ARGH!

I know they are on a white background, but i am sure you can understand my surprise to see such utter crap in Peer reveiw. I had to go check to se if i had selected Playtest instead.

I was disgusted, but more was to come. They are games created by a dev named Aab. Yes that is right. some crapola name to get them further up alpha sorted lists. It may not make any difference, but it does to me. They are trying to game the system, even if they are not doing a good job of it. so then i had to click into them to see what other horrors they could hold. I also wanted to see that they were suffering from noone reveiwing and passing them.

I was wrong... They were both at 50%, though not a single person in the individual games forums had posted that they had passed it. How odd that seems. Maybe they only have 5 friends or accounts (enough for 50% maybe?). I was also shocked to se the amazing and none verbose descriptions of these games.

TriangleReversi   AvatarViewer

Yea, that's right. This crap is gonna get onto Xblig, my beloved Xblig. further bringing down it's rep. Why is this dev doing this? Maybe they think they will make some dosh. I however sure hope they don't and then they go away asap.

That's enough of me moaning about aabs. I was gonna write about the games i tested, but BAH Humbug instead!

Laters