Saturday 23 December 2017

Bananna Pi M2 Zero

A very interesting board indeed, its a zero in its layout, but this time a quad core.. Sadly only a Mali 400Mp2 but it should out perform the Raspberry Zero.

As always though the software isn't here yet, Armbian does a fair attempt at getting it up and running but terrible graphic glitching makes it useless for any kind of graphic systems, and it GLMark2-es seemed to think it had OpenGLES3.0 GPU on board, the Mali 400's are strictly ES2.0 which probably explains why it wasn't able to complete the tests,though it did get quite far into them, though very slow

I'll do an update/upgrade cycle on it from time to time to see if there's any improvement, but for now this is a non starter for our gaming purposes.

Also interesting to note it gets bloody hot, I put a heat sink on it, wouldn't want to run without.





edit... ok I found a slightly older verison of the Armbian/ubuntu test build which installed at 720p and it is much more stable, after update and upgrade it is quite stable, no glitching and GLMark2-es2 worked better, showed as Mali400 and ran all the tests, but it reported a really crap score of 52, which is feeble compared to a Raspberry, even the Zero.

Not sure what to make of this, its quad cores make it a great unit, its crap graphics make it pretty useless, but functional barely. I guess as with all maker boards it depends what you want to do with it, if you need a small cpu system, its amazing.

Wednesday 13 December 2017

Hmmm bit too early for the UpCore

The UpCore has given me a few issues, not least of which was a reluctance to actually boot. But a request for help on the Up forums eventually yielded a solution, turn it on and off again...basically.

Removal of the battery for 30 seconds or so, and reconnect, saw it finally boot up.

But oh boy its slow. It also had no GL or GLES drivers, or GDB on board, though these were easy enough to install, and getting a buildable config was fairly painless.

However, despite that ease it has been incredibly reluctant to build the test map demo. The build refused to build with a -j4 settings despite being a quad core, hyper threading capable machine. It runs out of virtual memory, no idea what causes that.

But annoyingly, but after trying -j2 and getting a hang I went to -j1 and finally we get a build...a very slllllooooow build

But sadly not to completion, it eventually chokes on the modelmanger.cpp file, which as far as I can see isn't anything out of the ordinary. After sitting waiting for it to move on it gives an unspecified error. Cancelling the build on the PC side does not do anything to cancel on the target side, which becomes crippled as its building.

I used the same config but with a different target address for the Up^2 and it built and ran just fine, though with the known issues with shaders. As the Core is built on the same principles as the Up^2 its all a bit suspect.

As this is so new, I suspect the issues is more to do with a very early OS and certain features of the CPU not being fully activated (like multicore).

I'll put it away for a while and come back to it in a few weeks to see if there is an improvement.

Friday 8 December 2017

Sorry been busy...but got a new board UPcore

No time recently to try out the new Nano Pi's but I am hoping to this weekend.

I also took delivery of my new 4core Atom 1.84Ghz Up core board. This has to be one of the smallest intel base machines you can possibly have, its only about twice the size of a zero, but is markedly more powerful. Actually in theory I think it can out perform the celeron powered Up^2 I got in August. Which despite some issues I had with shaders put every other board I have to shame.  I can't wait to try it out.

I was hoping its eMMC memory would come ready configured with an OS to let me boot it up and try it but it needs a bit of TLC before it will show me something, as one of the 1st owners of these, there's really not a lot of content yet. Powering it up, gives me nothing at the moment, so I assume its blank and the eMMC needs flashing. But not quite sure what that is all about at the moment.

I will get on it after I've tried out the nanoPi's.
Its a nice little board, though the lack of Network port might be an issue, but hopefully my trusty usb extension with network port will work ok.


Sunday 12 November 2017

ohhhh this might be good

Friendlyarm are one of my favorite companies, they are pushing a lot of boundaries with their range, they don't always hit it, but at least they try. Their old M3 and T3 boards are amazingly fast, though of course suffer a little from very old GPU's. But that power might make it possible to do more than a few things to ease the load on the GPU.

That said, the GPU is at a disadvantage anyway because of the usual, lack of drivers. So I'm really excited to see 2 new revisions of the M2 and M3, now packed in the smaller Fire3 and Fire2A format.
Allegedly with GPU acceleration.

If thats true then this little board could be one of the fastest machines under $40, graphics won't be a massive change, but it will be fun to use. I've got both the models heading over to me. I'll see how they stack up. I "think" the chip is a Mali 400, it might not even be  an MP2, quite possibly a single...ugg....but 8CPU cores, if they don't burn up and throttle will be wonderful.

Friday 6 October 2017

Vim2 and OrangePi Prime

Got both of these in the last few days
Can't really say too much about them yet, the Khadas VIM2 is in theory a monster system, with an Octocore CPU and Mali 820 (not sure of the cores). It comes supplied with Android and I have to say its nippy as hell. I will try install Ubuntu, and give a report on its driver availability.
The VIM2 forums though are awful,  very hard to find anything so it might be a while before I get to it.

The OrangePi Prime is a Allwinner H5, with a Mali450. no idea if the previous post notice about GPU drivers will apply to this 64bit system. Of course Debian as supplied had no drivers, and attempts to install them went a bit off, so I'm having a go at installing Ubuntu.
I will say though while using it the Debian install was quite light but noticeably nippy.

Ubuntu installed, it does have GL, but no GLES2 libs so I installed Mesa, however GLMark2-es2 installed and runs it, but as usual it appears to be software emulated as its painfully slow, with a pitiful score of 15, oh well I'll see if the Allwinner gpu fix works on it.


Tuesday 26 September 2017

oh finally

this is very good news


OpenGLES will now be available, in the near future, on Allwinners range of SOC.

There are a lot of SBC's using Allwinner chips, and with this announcement we should start to see new versions of Linux with OpenGLES available. That will be a massive boost to anyone doing graphic coding.

This mainly impacts on the Orange and Banana range, Nano/FriendlyArm tend to use Samsung chips so it wont' be a direct benefit for them, but who knows it might encourage them to allow release of the binaries.




Monday 25 September 2017

NanoPi K2

Friendly arm, have very kindly supplied me with a couple of K2 boards along with cases/heatsinks etc.
I must say thank you to them for sending them as I was keen to try it out.

This is potentially a very nice machine, an Amlogic S905 powered quad core with Mali 5 core 450 GPU, Essentially the same basic layout as the Odroid C2 which I managed to get working but I think the drivers are emulated so it was curiously slow.

However at the moment the K2 is limited to an Android system, and a fairly basic one at that, and I can't find developer options to allow me to access it. I will have to wait for some more uptodate images, or a working Linux image, probably from Armbian.

Running in Android it is quite nippy, but a few problems exist, it does not seem to connect to the internet via a cable, and sadly Kodi, its main reason for existing, crashes and refuses to run. So....well that's a bit upsetting. I will try it on a wifi later.

If we get a linux that works, I will review it again later, for now, it has to exist only as a potentially cool board that should trounce the RPi, but may again be let down by the lack of drivers.

Sunday 17 September 2017

Machines coming on line

As I've noted a few times I have quite a collection of different machines, but the fact is not all of them are suitable for games. Quite some number for example do not have their OpenGLES libs in place and Mesa does not always manage to replace the missing parts.

I've only done a few basic test systems on most of my machines, and looking back now its clear that some need a bit more effort. But I have got a multi config demo up and running now, and it's very cool to plug in a machine, make a few small alterations to a config and get it working.

I will be posting a list of all machines that are compatible, out of the box, and which need some extra attention.

So far I've managed to get all the Raspberries all up and running with no problems at all but also the,
NanoPi M1 // not a hard one
Odroid C2 // really easy but slower than I expected, I might need to look into that
Up2 fast but some issues with shaders to research
Odroid XU4  // had some strange issues and need to add  -DGL_GLEXT_PROTOTYPES to the cc flags but once done its up and running and fast as hell
OrangePi PC2 needed a bit of coaxing but in the end does work, but software emulation of the GPU is happening and its very slow.

But also failed miserably with a few others.
Banana Pi M3.. no graphic drivers
NanoPi M3 no graphic drivers (shame as its bloody fast)
NanoPi M2 No graphics
Orange Pi Pc
BananaPi M1(original) hard work, tried multiple things, but code compiles,it just refuses to run,


Nearly
Tinkerboard... I don't have a 3A unit here, so will have to pick one up.
Olimex A20 Lime... couldn't get it to boot into linux...grrr lots of time wasted

Tuesday 12 September 2017

Streamer must wait

The new streamer I got to grab video images, hopefully in real time with coding, failed.. The drivers simply don't work, and there seems to be no way to get new ones, the website referenced in the manual is down, I don't know for how long


oh well the USB one is quite usable, I just need to edit things a bit more carefull.

Book proof reading/fixing is almost done, the publisher will get a final version this week, and then its all up to them. I hope.....

Thursday 31 August 2017

testing video

here's a short video to demonstrate a simple Line of Sight occlusion system.
I'm just trying out my video capture unit at the moment, no sound as yet though. I need to find a small mic to put sound in it


Since there's no sound, I'll explain, each character becomes the target of a ray cast from the player (which would normally hold the FPS camera), if the Ray hits the character it is white and the character displays. If a wall gets between the the character and the ray, the ray turns red to indicate it hit a wall and the character is not drawn, saving quite some amount of GPU time.
One character is actually being detected but not displaying, I'm not actually sure why, some small bug somewhere I suspect.

The system is pretty basic, and only a single ray is firing, for total accuracy you would fire at the extreme corners and only exclude the model if all 4 rays hit the wall. But ray casts on this are not cheap, so if you can get away with one, its better.

The ray is always being cast wherever you are in the map, to indicate that the draw system for that model is indeed still running but not drawing. A Distance or Frustum cull would probably also be used before a ray cast was done to reduce the amount of maths being done.



Wednesday 30 August 2017

Something different

And maybe misjudged, but I saw this on Gear Best and thought I'd give it a go.

Not an SBC in the usual sense, though it is in fact, but a TV box, designed to be used for KODI and other streaming systems, but at its heart it is a pretty nice SBC


Its an Octocore Amlogic S912 which in itself is impressive, but also its gpu is an even more impressive  ARM Mali-T820MP3  . Now that's a very modern GPU capable of ES3.2 and Vulkan coding, I only have a few units that can do 3.1 so 3.2 is quite a leap.

Its got 3GB RAM and 16GB eMMC as well as the usual connectors, and has dual boot Android and Linux, so....in theory at least, drivers permitting I should be able to code this, and if it pans out it could be one of the more powerful units in my collectiom. 
Or it might indeed just be a small Kodi or streamer box to sit behind my TV. We'll see I am curious to know if I can extend my remit to include such tv boxes, as progammer learning tools.

In keeping wiith my under $100 mantra it only cost $82, with a remote as well,  including the postage..there might be some duty when it arrives but we'll see.




Friday 25 August 2017

New board, feel the power

Well this is going to be a fun new toy,  today I took delivery of an Up2 (squared) board which I backed in a kickstarter some time back.
This will be my 1st intel based SBC so Im very curious to see how it performs. I went for the cheapest one to try to stay under my 100 euro limit, its actually 102 euros, but that includes the postage.

1st impressions are its a bit of a beast, the heat sink on it is massive and makes it quite heavy but it is passive, no fan, so I'm hoping it does its job an keeps the chip cool.
The cheaper version I went for is a dual core celeron 2GB memorywith 16GB eMMC its about twice the size of a RPi, and the free case I got with it, though a little loose keeps it all in a nice box.

I have a couple of different power units but this one needs 5-6 amps an a wide barrel jack, In think I have one here, if not I can get one.

I will try to do it this weekend, if not I will have to put it away for a bit.

But I am quite excited, on paper at least it should be considerably more powerful than any other board I have, even the Odroids, Its GPU has I believe 12 cores/pipes that will reall give graphic programming a boost, but I will be amazed if there are drivers easily available, but we'll find out soon.




Thursday 24 August 2017

Video's soon

I've ordered a nice little HDMI video grabber which should arrive soon, so that I can take videos of the games and demos running on the various kinds of Pi. Should be more instructive than the simple screenshots. I couldn't find a software solution to it, like rasp2png but for video, so went the hardware route. I hope it works ok.

I'll post a few here, most mostly they will go on the scratchpadgames.net website in support of the book.

Which is still in editing mode, some friends have been proof reading it and finding lots of missed spelling and grammar mistakes. So far nothing technical, but I'm sure there are a few errors that will slip past.

As well as the editing I'm getting the source code, clean and ready for download, and will then get the final polished versions of the games up and ready.

I must confess I'd lost a lot of my enthusiasm in the final weeks, getting the book out, there's still much I wanted to put in there but page count limits just stopped me dead and it became more of a chore to limit the content than to write it,. But setting up the website is giving me back some of that fun as I transfer more complex topics to the site.


Tuesday 8 August 2017

Done....kinda

The book is DONE!!!

well done in the sense that I don't think I can add any more to it, now we start a short editing phase, where I fix things that are wrong, or have become disconnected due to moving things around. 
Also some screen shots need retaking as pasting them into word resulted in a resolution reduction but as far as writing and content are concerned Im done.

If I am 100% honest, I wanted to do more, but where do you draw the line? Its probably going to be 850 pages when it hits the shelves,  I can't detail every single type of game.. I have literally cut, 300+ pages of game listings and stuff I was going to use,  and will post somewhere on my website when I clean it up. but the book needs to be readable, and 1250+ pages would be too much.

I start at hello world and by the end the reader should be able to create any 2D game they can dream up and be able to do a fully immersive 3D game and be capapbe of doing more if they get the foundations right.. I hope they get it.


Its been a long trek, and an interesting one, I discovered a lot about SBC's the Raspberry in particular, its such a fun  machine to code I just don't know why people are not pushing their games on it.n so much more satisfying to see your games take shape with your own efforts rather than an engine.

Maybe the book will make a small change there.


Normal updates on new machines and some game samples will now resume.

Tuesday 1 August 2017

The next day or so..

The book will finally be done, its been one helluva ride, and right at the death I've had a major unfixable problem with a model (need the artist to fix it) but I have been completing the write ups on some of the more technical aspects of the code

Physics and terrain  collision for example


Here's an example of a tile based environment with full terrain collision in place, and a god camera to view it all. It chugs a bit since the wireframe draws are painfully slow, but its a debug option.

Animations, culling, lighting, cameras,  shadows in shaders, and other bits all done so I'm just waiting for a model fix to finish it all off.

Its a lot of work, perhaps I didn't get quite the gameplay in there I  wanted but thinking about it, explaining how to write a particular game wasn't the idea, explaining how to write multuple games using the techniques explained makes more sense.

I hope this model I need gets fixed asap, or I will have to find an alternative way to explain a final point...or scrap it and wrap it up.


Sunday 25 June 2017

832 well its still not too bad,

Got the chapter about light in place now and the shaders for vertex and fragment lighting so thats pretty good. a bit more chopping done but no content really lost, I just noticed I was repeating myself a lot

here's the results of the vertex lighting shaders. 1st Screen shows the left model with no light, and right with light.




2nd image has both lit and you can see the lack of light on the belly of the upper ship.


Wasn't at all hard to do, but a bugger to explain simply and maintain a render system that could draw with or without light, thats not normally something I would do, but not to worry it works.



Saturday 24 June 2017

830 they'll kill me a bit less

Shunted some stuff to the Website, and rewrote some overly waffly bits...still more than they want but...hey...


Monday 19 June 2017

870pages my publishers going to kill me

well he's going to anyway for being so late. But I'll eave the edits to them.

Another good day done, I got more of the basic object defines in place to allow the physics code to be visible. Not without issues of course, there are always issues! But despite astonishing heat, and totally useless cooling in the office, I managed to get quite a bit done today, and write it up and endit a few other sections.

Back to work tomorrow though so not going to be able to do much, hopefully today's work will allow me to target the things I need to get done in the breaks.

I will get the MD2 models and lighting system sorted next. They are all working find, but the code is rough and needs a clean up, and a very slight introduction in the text.


Please be over soon...please! I'd like to enjoy this heat, sitting in my back garden with a pitcher of Margarita's not slaving in a hot room in front of 3 screens



Sunday 18 June 2017

is it done yet?

Oh no....but so close, I just wrote up in my main  blog how this last few steps are proving so hard, every thing that can go wrong goes wrong. Small simple tasks I expect to do in 5 mins take 5 hours and so on, its as if nothing I can do will actually work on a Pi.

But that said, I've pushed and pushed and pushed and the last few bits are almost done.. I've  written all the code to create the last few projects, apart from the implimentation of some physics and the rendering of the physics representation objects...that should normally only take a day at most....so expect it to take a month.

groan.

On the plus side, despite all this hassle the actual quality of the projects on the Pi is really amazing to see, its very clear to me that the machine simply isn't being pushed by people and the rather poor quality of games and graphics on offer can easily be exceeded.  So I really do hope my now massive (850 pages)  book encourages people to give these little beasts the attention they deserve it can do so much when its actually coded at its native level.

Ok enough posting...back to work.. No new SBC's bought recently, a few new Nano and Banana's have hit the market that are worth a look but I'm not going to get into them till the book is finished. The Tinkerboard is still the most exciting and its had a few updates to its OS, but still no drivers yet, I won't touch it till it has ES3.1/2 drivers.

Oh and the website is up now at Scratchpadgames.net There's nothing there yet, but I'll start to populate it soon.

Onward and downward, or at last try for a shallow attempt at an upward direction

Tuesday 6 June 2017

And another Giant step

The Skyboxes are working again, transferred from the old projects, and with the model loaders all working as they are supposed to we're up and running again.
the Space and Race games (same framework, different complexity) can now continue

and to add to that, a very serious and strange random bug has been squashed, and the camera is now functioning just fine, its all starting to work again...phew.

Finally moving again

After all the hassle with the car model I was looking forward to moving ermm forward...sadly the car was only 1 of my issues, I also have terrible problems with my skybox, until I realized I was not following the actual advice I put in my book. The part where I said, "always indicate to the user when you have a failure to load data".........yup, I hadn't put the textures in the transfer list so they never got sent to the target for building, so they never got loaded and I didn't have an error notice in the code!!!


Even very old game coders can repeat very basic errors.

And it can suck up time in ways you just can't image. Anyway, the car and skybox are in place again, now to get the track running and activate the physics.


Thursday 1 June 2017

WOW why do the simple things go wrong at the end!

I've had a nightmare couple of days trying to use a particular model in my race demo game. One of the last real coding tasks I had to do.
I wanted to use it for a number of reasons, not least of which is its a very cool model. which also comes with a normal and specular map.

But try as I might I could not get the damn thing to load, There were a range of small issue that stopped it.
The Jpg files didn't load correctly for 1, the obj references a texture on the makers PC, the 1st shape was a convex hull with no texture, there is a transparent window section, there's multiple detail shapes that don't need to be rendered (which is useful to explain occlusion)

Somewhere along the line, the jpgs and or the texture coords got mangled.. though they worked fine in other model viewers, TinyObjloader wasn't having it.

But after 3 days of exporting, in different packages, reducing the vertex counts and a range of other things.. I got it to work.
Behold, a 20,000 face model,  Though I don't have a suitable shader for the windscreen at the moment



Saturday 27 May 2017

Spell checking done

Yes I know, still not done, a few soft deadlines missed and 2 hard ones missed, but circumstances totally beyond my control have meant I just have not been able to get the last few sections up and running.

And a collegue  pointed out that my removal of an intro to physics, for space reasons,  left the remaining sections poorly coded and not at all up to standard, so I have restored but re-written the physics sections and am working as fast as possible to restore the code back into the final demos. Its worth the effort though the games are working much better with a physics engine, despite the performance hit.

Just as I was ready to put it all together, I ended up being stranded in London after visiting for a conference, and losing my passport, which was a major pain, but also allowed me to do a full read through and review, find a few more errors and get them marked for fixes which I've been doing since I got back home,

The Website is up and ready for content, www.scratchpadgames.net Its only in a holding mode at the moment but it will house all the demo code, some extension lessons and answers to some of the questions posed in the book.  I'll also add a small forum  to it when I work out how :D At least that does not have to be done until the book is launched sometime after November!

Its actually listed on Amazon now, ready for pre order, which of course emphasises the need to get the damn thing done https://www.amazon.co.uk/Fundamentals-Game-Programming-Target-Based-Development/dp/1498788742/ref=sr_1_1?ie=UTF8&qid=1495885321&sr=8-1&keywords=brian+beuken
The price still worries me though...but no doubt it'll come down when the try to unload the unsold copies lol.

I also did the 1st full spell check on it last night, took 4 hours lol....I can't spell for toffee, being slightly dyslexic, That's one of those silly but essential jobs that I need to do a few times just to make sure. Its complicated by the fact there is a lot of code in the early sections that will confuse any spill chucker. It will no doubt go through another spill chucker to turn it to american English when the publisher gets it.

I need to get back to it now, I have a particular model in the game I got online, its lovely but it is giving me a  lot of headaches to display..hopefully after a break writing up the other games when I get back to it it will be cool.




Sunday 16 April 2017

4 games in one :D

I've had to do a few re-writes to cut pages and reduce the amount of code in the book,  this is mostly done by letting 1 project flow into another. Rather than having,as I did, separate projects. So I've build a simple tile display system, which in turn becomes a top down shoot and chase game, then a single screen platformer, then a scrolling platformer, and finally a scrolling shooter.

While at the same time introducing collision, motion, logic, attack waves and for fun a bit of parallax scrolling.

All using the same graphics :D


here's the shooter


Its reduced the page count quite a bit, and It think made the flow of information that much cleaner...I'm,doing something similar for the 3D games in the book too. Still maintaining the style but allowing 1 to flow into another.


Wednesday 12 April 2017

heavy edits ongoing

Been working on converting a single screen platform game like this



into a multi screen scrolling game...these were originally 2 separate projects but space has become critical in the book so I needed to put the two projects together and show how to turn the 1st into the 2nd with the aid of a couple of simple concepts.

Saved quite a few pages and I think it makes the experience a lot more fun.



Tuesday 4 April 2017

some new toys

Strange formatting warning...cut and paste from Facebook, but lots of spaces added! 2 oops 3 new Orange Pi's picked up from post drop, had to pay some duty on one...almost as much as it cost, ironically the cheapest one...but still

crazy silly little Orange Pi Zero Plus 2 512Mb machine with onboard 8Gb eMMC memory.... this thing is tiny, has wifi on board, but works better with an expansion board to provide usb connections,
And an Orange Pi Win Development Board A64 Quad-core,
Which even as it was arriving here, was replaced by a new 2Gb version, but I won't be buying that yet.
These OPi's keep coming out faster than I can collect them... Shame the software support for them is utterly terrible. Going to experiment with Armbrian on these soon and see what they can do.
Oh I forgot last week I also picked up a "normal" Opi Zero H2 model with an expansion board and tiny case....crazy!
I'm going to start getting all these various machines hooked up to a central server system and have them be my household controller units....or just write shit games on them.
But for now I have to continue with the book, so they will go on the shelf for a couple of weeks.

Monday 3 April 2017

Definitely the home straight

Last week I hope of book writing, mostly editing and documenting some code.

Still got a 100 pages to try to chop out but I think I will manage it then I can hand it over the publisher and get some rest.

In mean time I just finished doing particles and emitters using a 2D CPU rendering sytem..

looks pretty


Thursday 16 March 2017

Damn its emulated

The OpenGLES3.1 (or anything else that I can tell) is not runnong on the GPU, its all being emulated a the moment, I guess we have to wait for a version of the os with proper libs in it.

But even so, I expect these to come on line and its sitll possible to code in 3.1 albeit slowly.
So yeahhhh but, only lower case for now.


Wednesday 15 March 2017

Got one.

Yeah I got my Tinkerboard at last. It is a very nice clean and tidy board,

After a bit of hunting following the ghost launch I managed to locate one in Germany, but it seems the supply is starting to ramp up and they are now flowing into the market.

The support site is still generation 1 and nothing by way of a community, but I was able to locate and download Debian. Sadly at work, using a Pi power unit it refused to boot, however once home with the same image and a beefier PSU it fired up in no time. It is nippy and responsive on the desktop, and after enabling its auto network, it went on line and happily performed and update/upgrade

It is a no frills standard Debian, it resized my sd card itself, and logged in to a sparse gui with nothing notable to report, except for one amazing features. It has OpenGLES3.1 libs on board. I’ve not had a chance to try them out yet, as I’m a bit busy but if this is indeed a properly implemented OpenGLES3.1 system, its hands down going to beat almost all the other SBC’s out there. I still want to see 3.2 on this thing though since ARM's site confirmed the 760 series are 3.2 capable


https://www.arm.com/products/multimedia/mali-gpu/high-performance/mali-t760.php


CPU gets mighty hot, so don’t try to run it without the heat sink, and as I say it needs a good power unit.

So far so good. I can’t wait to get programming some OpenGLES3.1 on it and see what it can really do. Sadly I just don't have time at the moment as I'm rushing to finish the book, but there are a few proejcts that will need testing on a non Pi machine... I wonder which ones I might use :D

Tuesday 14 March 2017

Editing begins and Tinkerboard on way

so while I am still doing some code fixes , I have finally made a start on the 1st edit of the book. There was/is a fair amount of repetition of some concepts, and some things I can easily hive off to the support site. So pretty sure I can lop a fair chunk of pages of the current 860 total.

Most projects are starting to firm up now, only thing giving me big headaches are systems that need internal environments, its a complex subject to discuss, even more complex to write about but hopefully the demo's I'm putting together will demosntrate it.


I also managed to find a site in Germany that had stock of Tinkerboards, I am very excited about this board, more than any other due to the potential for mass market it has. Its still rather poorly set up, but its only the 1st generation. Can't wait to report on it when it arrives

And I am really pleased to report a lot of my students have been having a really good time with RPi and developing some concepts I outline in the book, especially the use of MD2 models and OBJ environments, which will allow for a lot of fun complex games with animation.

Here's a screenshot of one of my students projects using an internet sourced MD2 based animated model and a Mario world backdrop, its kind of a funky mash up but so nice to see what they can put together.its a shame I can't grab video from the Pi, seeing a model animate and move around an enviorment so smooth is really sweet.




Friday 10 March 2017

So behind...but moving on now

Arrh tech issues, the bane of any coders life. I've had a few.
Book writing is going fine, book coding has been hitting bottle necks right left and centre, the worst of them being an instability in Visual GDB's debugger. It just became impossible to debug code as every break or single step would cause a network freeze.

It took a bit of hunting down but I finally found it, the Rasbian version of GDB is outdated and Visual GDB causes it to crash sometimes.... which is a pain.... You might expect the VisualGDB people to fix that but they don't accept responsibility for 3rd party issues...which is kinda understandable but frustrating.

Anyway, after much research, and digging into Linux (uggg I feel dirty) I worked out that replacing the GDB on the Raspberry cures the problem totally, its not an easy fix but it worked.

I now have a fully functioning debugger again. yeahh

And here's a screen show of an alien armada to celebrate.



I've also got MD2 animated models working and shadow and light systems all working, these were the tricky techy bits I really needed the debugger for, and now that they are done I can kinda forget about them and get back to the simple game

There are still lots of strange tech issues, why is the frame rate locked at 60fps for example, making delta time useless..but I'm asking around and will get the answers.


Wednesday 15 February 2017

Where'd it go?

Well the Tinker Board seems to be suffering from a serious case of premature release. The lack of support has caught Asus out and I hear that very few people have managed to get their boards up and running.
The only supplier in Europe I know of has now withdrawn it from their catalogue.

I do hope that does not sink the chances of this very promising board, I will keep my eyes open for it

Book news....
not had any for some time have we, vurrentl y 825 pages, it was upp to 900 but I edited it down a bit. Its coming to an end (again), a couple of projects have proven to be harder to document than I expected, and I had to cut the scope of them down quite a bit but its coming

I'm due to hand over at the end of Feb, I don't think there's a real chance of that, which might annoy the publisher who's been incredibly support over my delays but I am pushing my luck now. But I do think it will be done next month (mid March), so hopefully I can post some screen shots and pics of the projects soon..


But to writing.





Saturday 4 February 2017

Asus Tinkerboard, a new dawn for SBC's?

Sorry to have been quiet for so long, been both working on the book and recovering from illness which in turn caused delay in the book, which in turn meant trying to catch up on the work....I'm falling very behind and have had a few extensions to my deadline but am doing my very best to get back into the groove again.

But I had to post something about this beast



This ordinary looking board may just be the most important new SBC ever released, for 2 important reasons,
1st, its from ASUS, one of the very biggest commercial electronic makers, these guys do everything from full Mother boards to tablets to Video cards...and video cards is kinda interesting.

Their release of an RX3288 power credit card board with a very soft and unheralded launch might mean they are checking out the market.

2nd the use of the RX3288 is in itself interesting, yes its 32bit, but its a very fast 32bit, More than capable of showing a Pi or even an Odroid its clean heels. But it won't be the outright champ, there are a few faster CPU's,
However it is running a Mali 760 series GPU....4 cores.  In theory...that means its capable of running OpenGLES3.2 making it the 1st sub $100 board to be so empowered.

If they are also able to use their experience of video cards to provide full drivers for this board, it will make this the most powerful SBC board available on the market if you are planning to do graphic applications, some of the 64bits will out pace the CPU, but none currently come close to a 760 series GPU.

The launch has been incredibly low key, only 1 supplier in the UK, and sadly they sold out before I could get one. What info I can find so far suggests that they don't really have the software side sorted out, and images are basic, and tellingly lacking in OpenGLES3.2 (or any) drivers....
But I hope that this takes off. If it does we might see all the other makes move up a level in GPU performance and might also see people taking more interest in coding graphically.

I'll let you know when I get my hands on one.

Back to the book