Matrix Games Forums

We put out a new Build #214 today. Many bugs have been squashed. A few more persist. We're working hard to iron these out. The scenario designers are going to be making some changes to accomodate the new features and behaviours and we'll be playtesting these to get them just right.

By way of example today I made some mods to the Assault on the Sauer scenario. This is one of the OTag scenarios - ie it starts on Dec 16 when the Germans launched the offensive. - and covers the assault by the 7th Army on the southern flank. We were having a devil of a job trying to get the US forces to stay in line but the new code changes that pre-allocate forces dug-in or better within the perimeter of an objective has done the trick.

This new code is a pre-allocation so it biasses the unit's allocation to the objective it starts within. But it doesn't preclude reallocation if other priorities warrant, such as when someone needs to be found to secure an undefended objective. Here's a shot from the game showing the effects. Note that one of the US Inf Platoons has been re-allocated out of its original position to secure another objective. ( Don't worry we've subsequently edited this scenario to ensure that objective is garrisoned at start. )

Also note that A.60 Coy cannot prevent the German's from infiltrating south past its flank during the poor visibility of the night. To redress this we're going to be substituting three platoons for that single company unit and deploying these in line to prevent that infiltration.

Looks great guys. Any updated ETA info? Somtime this Fall I would imagine?

Thanks

ETAs are overated! [;)] It all depends on how long it takes me to iron out these remaining bugs and for Paul to finish off his UI fixes. These could be done this week but may take a lot longer. I just cannot say for certain. Once we have a robust game it will take another month or so to wrap up the tutorial movies and documentation.

Because waiting on this every single day would have been too painful, my strategy has been to check out some other games AND WORK TO FORGET THIS ONE until the day it comes out. And the very instant I see it available, I will BUY. No questions asked. I am just amazed at the leap of progress it appears Panther Games and Dave will have taken on this one.

My ONLY complaint with what I see so far is that I still can't tell what is going on with the elevation lines in the woods. I really feel strongly something needs to be done about this. However, AFTER the initial release! Or I will go crazy waiting![&o]

ORIGINAL: Arjuna (9/19/2009 3:35:02 PM)

We have it slotted in with Matix for a pre Thanksgiving release....subject, of course, to us being able to sign off by the end of October.

Just thought I should pass Dave's statement which slipped out in the COTA forum (Scenario design section), harhar.

Thanks for the progress info....

ORIGINAL: barbarossa2
My ONLY complaint with what I see so far is that I still can't tell what is going on with the elevation lines in the woods. I really feel strongly something needs to be done about this.

You may notice from Dave's more recent screenshots that the new woods graphics are more transparent than they were in COTA or HTTR. That is all we could do to help with the elevation differences for now.

ORIGINAL: Arjuna

ETAs are overated! [;)] It all depends on how long it takes me to iron out these remaining bugs and for Paul to finish off his UI fixes. These could be done this week but may take a lot longer. I just cannot say for certain. Once we have a robust game it will take another month or so to wrap up the tutorial movies and documentation.

Any chance of a mini-update??

Rob.[:)]

SITREP Friday 25 September 2009
I'm not sure if I mentioned this previously but one aspect of the Bulge that has a serious impact on the German ability to prosecute their plans was there inability to keep their motorised formations fueled. We researched the amount of fuel they were able to feed into the area of operations and used this to seed the resupply schedules. Let me tell you it ain't much. Consequently, German units run out of fuel quite a bit ( as they did historically ). Now under the COTA system if one unit runs out of fuel and stops and it's moving in formation, then the whole formation stops. An emergency resupply request is sent and the force waits till the resupply arrives and then they all head off together again. Now in COTA this worked fine because there was invariably more fuel at the depot supplying the force and the turnaround time was only a couple of hours.

But in BFTB often there won't be any fuel at the depot and the force as a whole could end up waiting a day or more for fuel to arrive. Historically, the Germans simply left the out of fuel unit behind and pressed on with the remaining force. Only when the whole force was down to emergency fule levels ( around 10% ) did the whole force stop - eg Piper at La Gleize. To be able to simulate this I have added new reaction, reasessment and planning code. Now a unit that gets down to 10% fuel will temporarily abandon its mission, request emergency resupply, advise the boss, defend inSitu and wait for the fuel to arrive. The boss will then normally regig its formation, hive off the refueler and press on. However, throughout the planning and reasessment code changes had to be made to split the force, with refuelers being hived off and issued defend inSitu orders.

I've made all these changes over the last month and it works quite well. Though, there was a swag of knock on effects throughout the scheduling code - one of the problems with a complex code base. I've been working my way through these over tha last week. I have a few more asserts still to work through. But I'm reasonably optomistic of knocking these over very soon.

Another issue that I have also addressed this last few weeks concerns attacks. When you issue an attack order without specifying an FUP, the AI usually develops three routes to the objective from the subject. One is a frontal or direct route. An FUP is determined along that route. For the subsequent routes, it uses an exclusion path. This avoids enemy firepower but also excludes all known friendly FUPs and objectives ( other than the current objective ). That way we avoid getting in the way of other attacks.

However, in the BFTB, where there can often only be one viable crossing point, we can end up with a situation where you have ordered an attack to secure the crossing point. It's going well. You order your reserve Bn to attack an objective on the far side of the river. You know that by the time this Bn processes its orders the initial attack will have finished and secured the crossing. Alas, the AI doesn't know that. As far as its concerned there is still an attack objective at the crossing and so it excludes the area around it. Consequently your followup attack gets abandoned as there is no valid route. I was finding this very frustrating and I'm sure you would too.

So I bit the bullet and implemented an aversion route rather than exclusion route. Now rather than excluding those locations ( FUPs and objectives ) it simply adds a very high cost to them. That way it will avoid them but go through them if there is no other option. This is a neat bit of code, originally written by Sos a few years back, but never implemented before. I was so impressed, I extended the aversion to include the other approach routes for the current attack being planned. Anyway it works a treat.

This game will really raise the bar for realism. I'm really looking forward to this title.

Thanks for the update Arjuna!!

This looks to be the best yet. I can see the aversion code (esp.for bridges) working extremely well in the 2 or 3 HTTR scenarios you previously indicated would/will/might be included in BFTB. Is that still the plan i.e. including some HTTR scenarios with BFTB?? Or perhaps I misunderstood, and they'll be in an expansion pack??

Rob. [:)]

No we won't be including any HTTR scenarios in BFTB. There's already 27 scenarios included. As I have said before, we will be reviewing future development options after BFTB. An HTTR expansion pack will be in the mix of options, but I'm not prepared to commit to that at this stage. The focus will be on providing data content sooner rather than later.

I have been head down tail up these last few weeks squashing bugs. We have an "autotest" system where we run the game continuosly. If an assert or crash occurs it copies the details to a log file. It also logs games that fully complete. Each day I make a list of the asserts and start working my way through them. At the end of the day I update the version of the game used by the autotesters and repeat the process. Two weeks ago we had a ratio of asserts to games run of 1:1 - ie 100% of the games asserted. We're now down to only 20% - ie for every five games run only one asserts and the others run through to completion. With a bit of luck we'll have ironed out all of them by the end of next week. I can then focus on the remaining UI bug and ScenMaker bug and then the tutorial movies.

My son, Sean, whose just finished university is helping me draft the stroyboard for the movies. So I should be able to dive straight into the movies by the time I finish off the coding. Our aim with the movies is to provide a quick and easy way for users to get into the game. At this stage we plan to produce several movies covering the tutorial scenario. We'll also aim to produce a "What's New" movie covering the new features.

How will those movies look, only in-game view or will you seed in some other things like animations, historical footage, etc.?

BTW that "What's New" sounds good, did you think about giving it "for-free" to the community?
I bet it would make a good advertisement for the game.

At this stage we won't be including any historical footage. The aim is to instruct on how to play not to set the scene. As to animations, that will depend on whether or not they are needed. We'll have to try it out and see first. The intent for the Waht's New is to make this publically available. In fact the aim is to include the tutotial movies and the WHat's New movie with the demo.

Great to hear the automated testing is helping.  I know it's a big up-front investment in time writing the scripts etc. but from what you are saying it's really giving you good test coverage and focussing effort on fixing instead of finding.

Kipper,

Bear in mind that autotesting is good at finding AI bugs but not UI ones. So we still need to have real humans ( as opposed to those fake one's [;)] ) playing the game manually to find the UI bugs.

ORIGINAL: Arjuna

Kipper,

Bear in mind that autotesting is good at finding AI bugs but not UI ones. So we still need to have real humans ( as opposed to those fake one's [;)] ) playing the game manually to find the UI bugs.

PG's dedication to only releasing stable and bug free (99.9%) games is impressive. That is another reason I will always buy your games on day one of release.