Captains!
Happy Friday and welcome to the weekend! This week I have a couple of small points for you- really just to bring you up to speed on where we are, what we are doing, and how things are going.
Before we begin: This post is primarily focused on giving you a real-time look at how we change development priorities in order to advance products toward release. Honestly the topic is a bit insanely dull to me, but many of you have expressed an interest in knowing how we handle changes behind the scenes, so here is a real-time look.
PMDG 737-700: Pending Update
This update cycle has a number of of fixes/tweaks and new capabilities added to it, but most of them are really small functionality items. The larger focus has been on continued upgrades to targeted areas of the flight deck visual fidelity and improvements in animation algorithms that have proven problematic for many users in the form of "hitchy" or "granular" or "low fps" animations in things like the yoke, column, throttles, etc.
We are happy to report that we have made really good progress toward smoothing these things out. To give credit where it is due- a couple members of the user community made observations about the "low fps" that got the investigative progress going in the right direction- and we found that animations tied to complex systems where we intentionally limit the computation speed in order to avoid flooding the processor with floating point computations is where things break down. So a reimagining of some key areas related to the measurement and passing of time and processor cycles was called for- and this is miserably complex work that is taking a bit longer than we had planned.
The good news is that the results are starting to show and we are relatively confident we will be able to apply this technique to just about every area of animation to smooth things out. It might take a few update cycles to get through all of it- but the first round of it is focused on the main offenders sitting right in your central workstation on the airplane.
So those of you who have participated in the conversations on this topic here in the forum, and those who have answered questions publicly and privately- please do pat yourselves on the back- nice job in helping us zero in on the core of the problem!
Oops, I did it again:
Anyone who has ever worked with computer animation will tell you that one of the hardest things to animate WELL is an object that spins really fast. Wheels, windshield wipers, fan hubs, and trim wheels. I hate all of them.
We spent a ton of time getting the trim wheel animation correct in the 737 prior to release- and then along the way someone (who shall remain nameless because he signs the paychecks <cough>) accidentally reverted that section of code to "original source" which then took the 737 trim wheel animation back to a version that was implemented in the 737NGX in 2010. And nobody noticed.
After seeing a number of complaints about it here in the forum, I finally got around to looking at it more closely and realized that it was indeed hopelessly broken... A day re-engineering it to approximately what I remembered it to be prior to breaking it in the first place- and we were ready to go. We issued a new build to our beta testers on Monday, thinking "good- we can get this fixed and push an update on Wednesday."
Then the beta team began to agitate about some problems in the animation and some really unwanted behavior in the sounds associated with the trim wheel. By Wednesday afternoon, Alex and I were working on this together trying to un-<censored> the trimwheel logic for the second time in a week.
We are almost finished- but need a few more days to hand it off to the beta team for testing and evaluation.
Since the trim wheel is central to the entire airplane- we have opted to sit on our hands with respect to a new update until Alex and I finish getting this un-<censored>. Or, to be more candid, until Alex finishes fixing the parts I have now broken three times. Nothing like having your development partner sabotage your forward progress.
PMDG 737-700 Update Estimate?:
We are currently hoping to put the update back into testing on Sunday, which will hopefully give us a validation to release the next update mid week.
Provided nobody breaks anything.
PMDG 737-700 into MS Marketplace:
We have held our fire on moving the 737-700 to Marketplace because we want to get a few more things corrected before we expand the user base via marketplace. Primarily, our concern is the incompatibility between our product and gamepad/twist grip/dual rudder axis controllers. This simply hasn't gotten the attention it needs yet- and we think it is ill informed to add the product to marketplace until we have a solution in place. I am desperately hoping the next update cycle (after the one in testing) will include a fix for these controller issues... Stay tuned for more.
PMDG737-600:
We have some exciting news with the upcoming 600 release: The beta team will be getting a version of the 600 over the weekend that has the new PMDG Lateral Flight Path Model that has been under development for two years. The new LFPM encompasses the entire lateral flight path process, including the flight director logic in order to give a far more accurate, more true-to-life lateral flight path model that has been sorely needed in our entire product catalog for some time.
The current expectation is that the 737-600 will deliver with this modernized capability, and the 737-700 will receive an update to build it into that airplane on the same day the 737-600 is released. (And then, obviously, it will be present in all products released going forward of this update.)
The new LFPM also has a fringe benefit of eliminating a long-standing problem in the vertical flight path caused by the fact that the older LFPM did not accurately compute the distance the airplane would travel in a turn, and this caused vertical errors to accumulate, sometimes becoming noticeable or causing the airplane to depart from the planned vertical path in ways that weren't expected. The new LFPM reduces these accumulated errors, resulting in an associated improvement in the vertical path, which has been waiting for the new LFP in order to improve accuracy.
Inevitably, our plan is to roll this new LFPM backward through our entire product catalog in Prepar3D as well. There is no defined schedule for doing this yet, as we are entirely committed to the MSFS dev cycle for the near term- but it is likely we will pop it into the P3D product lines in intermediate update cycles along the way. It is hard to predict a schedule for this until we get clear of the 737 release cycle at least.
PMDG 737-600 Release Plan?:
We decided to test the new LFPM in the 600 because doing the testing in the 737-700 would potentially mean some disruption to our ability to update that airplane rapidly if warranted. Putting the LFPM into the 600 allows us the freedom to give it adequate testing time where the worst that might happen is a bit of a slide in the release schedule for the 600. Currently we are looking at adding a few weeks to the 600 development in order to get this testing wrapped up effectively.
I'll have slightly firmer guidance on the 737-600 release plan next week after we see the next version in testing. I
Since one or two of you are wondering: None of this affects the dev cycle of the 737-800, which is progressing apace.
PMDG 737-600 in MS Marketplace:
We have been asked by a number of users if we would consider releasing the 737-600 in marketplace more-or-less concurrently with the release on PMDG.com. We are moving in the direction of saying yes here- but there are a couple of items we want to resolve across the 600 and 700 before officially declaring this a sound idea. I suspect the latency between a PMDG.com release and a Marketplace release will be smaller on the 600 than it is on the 700... and likely smaller still for the 800, etc.
PMDG Universal Flight Tablet:
Work continues here. This week we are integrating navigraph charts, and also refining simbrief functionality. It is still too early to predict when this will slot into the 737 release cycle, but the team working on this aspect of the airplane is motivated and making incredibly good progress. I suspect holding off on this technology piece until after release of the 700 has given them a chance to really refine what it is we are accomplishing with this piece of the product line- and it is shaping up to be something truly wonderful.
Far nicer than the clunky first generation EFB we modeled in the NGXu.
Alright- that was just a small update. We hope you all have a nice weekend ahead. Please do send in pictures of your weekend 737 flying adventures. Or- if you are out there flying for real- tell us about that too! Perhaps I'll share a flying adventure story with you as well here this weekend. Lets see how my wrestling match with the trimwheel goes.
Thanks everybody- enjoy your Friday!
Happy Friday and welcome to the weekend! This week I have a couple of small points for you- really just to bring you up to speed on where we are, what we are doing, and how things are going.
Before we begin: This post is primarily focused on giving you a real-time look at how we change development priorities in order to advance products toward release. Honestly the topic is a bit insanely dull to me, but many of you have expressed an interest in knowing how we handle changes behind the scenes, so here is a real-time look.
PMDG 737-700: Pending Update
This update cycle has a number of of fixes/tweaks and new capabilities added to it, but most of them are really small functionality items. The larger focus has been on continued upgrades to targeted areas of the flight deck visual fidelity and improvements in animation algorithms that have proven problematic for many users in the form of "hitchy" or "granular" or "low fps" animations in things like the yoke, column, throttles, etc.
We are happy to report that we have made really good progress toward smoothing these things out. To give credit where it is due- a couple members of the user community made observations about the "low fps" that got the investigative progress going in the right direction- and we found that animations tied to complex systems where we intentionally limit the computation speed in order to avoid flooding the processor with floating point computations is where things break down. So a reimagining of some key areas related to the measurement and passing of time and processor cycles was called for- and this is miserably complex work that is taking a bit longer than we had planned.
The good news is that the results are starting to show and we are relatively confident we will be able to apply this technique to just about every area of animation to smooth things out. It might take a few update cycles to get through all of it- but the first round of it is focused on the main offenders sitting right in your central workstation on the airplane.
So those of you who have participated in the conversations on this topic here in the forum, and those who have answered questions publicly and privately- please do pat yourselves on the back- nice job in helping us zero in on the core of the problem!
Oops, I did it again:
Anyone who has ever worked with computer animation will tell you that one of the hardest things to animate WELL is an object that spins really fast. Wheels, windshield wipers, fan hubs, and trim wheels. I hate all of them.
We spent a ton of time getting the trim wheel animation correct in the 737 prior to release- and then along the way someone (who shall remain nameless because he signs the paychecks <cough>) accidentally reverted that section of code to "original source" which then took the 737 trim wheel animation back to a version that was implemented in the 737NGX in 2010. And nobody noticed.After seeing a number of complaints about it here in the forum, I finally got around to looking at it more closely and realized that it was indeed hopelessly broken... A day re-engineering it to approximately what I remembered it to be prior to breaking it in the first place- and we were ready to go. We issued a new build to our beta testers on Monday, thinking "good- we can get this fixed and push an update on Wednesday."
Then the beta team began to agitate about some problems in the animation and some really unwanted behavior in the sounds associated with the trim wheel. By Wednesday afternoon, Alex and I were working on this together trying to un-<censored> the trimwheel logic for the second time in a week.
We are almost finished- but need a few more days to hand it off to the beta team for testing and evaluation.
Since the trim wheel is central to the entire airplane- we have opted to sit on our hands with respect to a new update until Alex and I finish getting this un-<censored>. Or, to be more candid, until Alex finishes fixing the parts I have now broken three times. Nothing like having your development partner sabotage your forward progress.
PMDG 737-700 Update Estimate?:
We are currently hoping to put the update back into testing on Sunday, which will hopefully give us a validation to release the next update mid week.
Provided nobody breaks anything.
PMDG 737-700 into MS Marketplace:
We have held our fire on moving the 737-700 to Marketplace because we want to get a few more things corrected before we expand the user base via marketplace. Primarily, our concern is the incompatibility between our product and gamepad/twist grip/dual rudder axis controllers. This simply hasn't gotten the attention it needs yet- and we think it is ill informed to add the product to marketplace until we have a solution in place. I am desperately hoping the next update cycle (after the one in testing) will include a fix for these controller issues... Stay tuned for more.
PMDG737-600:
We have some exciting news with the upcoming 600 release: The beta team will be getting a version of the 600 over the weekend that has the new PMDG Lateral Flight Path Model that has been under development for two years. The new LFPM encompasses the entire lateral flight path process, including the flight director logic in order to give a far more accurate, more true-to-life lateral flight path model that has been sorely needed in our entire product catalog for some time.
The current expectation is that the 737-600 will deliver with this modernized capability, and the 737-700 will receive an update to build it into that airplane on the same day the 737-600 is released. (And then, obviously, it will be present in all products released going forward of this update.)
The new LFPM also has a fringe benefit of eliminating a long-standing problem in the vertical flight path caused by the fact that the older LFPM did not accurately compute the distance the airplane would travel in a turn, and this caused vertical errors to accumulate, sometimes becoming noticeable or causing the airplane to depart from the planned vertical path in ways that weren't expected. The new LFPM reduces these accumulated errors, resulting in an associated improvement in the vertical path, which has been waiting for the new LFP in order to improve accuracy.
Inevitably, our plan is to roll this new LFPM backward through our entire product catalog in Prepar3D as well. There is no defined schedule for doing this yet, as we are entirely committed to the MSFS dev cycle for the near term- but it is likely we will pop it into the P3D product lines in intermediate update cycles along the way. It is hard to predict a schedule for this until we get clear of the 737 release cycle at least.
PMDG 737-600 Release Plan?:
We decided to test the new LFPM in the 600 because doing the testing in the 737-700 would potentially mean some disruption to our ability to update that airplane rapidly if warranted. Putting the LFPM into the 600 allows us the freedom to give it adequate testing time where the worst that might happen is a bit of a slide in the release schedule for the 600. Currently we are looking at adding a few weeks to the 600 development in order to get this testing wrapped up effectively.
I'll have slightly firmer guidance on the 737-600 release plan next week after we see the next version in testing. I
Since one or two of you are wondering: None of this affects the dev cycle of the 737-800, which is progressing apace.
PMDG 737-600 in MS Marketplace:
We have been asked by a number of users if we would consider releasing the 737-600 in marketplace more-or-less concurrently with the release on PMDG.com. We are moving in the direction of saying yes here- but there are a couple of items we want to resolve across the 600 and 700 before officially declaring this a sound idea. I suspect the latency between a PMDG.com release and a Marketplace release will be smaller on the 600 than it is on the 700... and likely smaller still for the 800, etc.
PMDG Universal Flight Tablet:
Work continues here. This week we are integrating navigraph charts, and also refining simbrief functionality. It is still too early to predict when this will slot into the 737 release cycle, but the team working on this aspect of the airplane is motivated and making incredibly good progress. I suspect holding off on this technology piece until after release of the 700 has given them a chance to really refine what it is we are accomplishing with this piece of the product line- and it is shaping up to be something truly wonderful.
Far nicer than the clunky first generation EFB we modeled in the NGXu.
Alright- that was just a small update. We hope you all have a nice weekend ahead. Please do send in pictures of your weekend 737 flying adventures. Or- if you are out there flying for real- tell us about that too! Perhaps I'll share a flying adventure story with you as well here this weekend. Lets see how my wrestling match with the trimwheel goes.
Thanks everybody- enjoy your Friday!
Danny Z.Cebis
.
Comment