Devnote Tuesday: QA focus

[Capt Snuggler]

Any chance we could get a peek at what @Porkjet is up to with the part revamp?

Edited September 1, 2016 by Capt Snuggler

[klgraham1013]

Guys.  @Porkjet is shy.  You're making him uncomfortable.

[TheDestroyer111]

3 hours ago, TheDestroyer111 said:

What the Laythe is all this QA stuff???

Do I have to spam the thread to make someone say something meaningful instead of pestering Porkjet? My question is still unanswered.

[Red Iron Crown]

5 minutes ago, TheDestroyer111 said:

Do I have to spam the thread to make someone say something meaningful instead of pestering Porkjet? My question is still unanswered.

QA is Quality Assurance, the first major phase of the testing cycle of a KSP release. It basically goes like this:

Developer testing (internal) --> QA testing (private, small pool of testers) --> Experimentals testing (private, large pool of testers) --> Public release

1.1 added another stage between Experimentals and Public release: Prerelease Testing. This gave much of the public access to builds of KSP before the update officially went live. It's not clear at this point whether that was a one time thing or if it is planned for this or other future updates.

[AlamoVampire]

I thought squad wouldnt do a prerelease again because of the backlash from last time? I mean after all only steam users got access which angered non steam users.. Or did i miss something on here about it going to all users who want it? Possible i missed that...

Edit ok ignore that didnt see the end of reds post

Edited September 1, 2016 by AlamoVampire

[Terwin]

21 hours ago, NathanKell said:

A bit more on the transfer bit that didn't make it:

To do it I took CrewTransfer and made it into a base transfer mechanic that both CrewTransfer and (now) ExperimentTransfer inherit from. This means that any mod can also easily inherit from it to make their own item transfer mechanic; all it takes is overriding what info is needed to start, methods to report whether a part is a valid (or semi-valid, i.e. can take crew but is full) part, and a method to handle when you click on the destination part.

This is making it sound like I will be able to transfer science data between pods the same way I can now transfer Kerbals(a boon for those who use a detachable ground level pod for Eve landers), also providing a way to transfer some but not all of the science data in a given pod to another pod(so I can keep a probe core stocked with one of each experiment I have not yet sent to my lab, letting me transfer out anything that I accidentally re-add after it was already researched.  This would also allow me to transfer experiments to a 'return' probe while the original probe moves on to the next target...  So many possibilities...)

I would assume that you put in some sort of 'one way transfer' mechanic so that I cannot store a copy of my gravioli readings in a spent goo canister...(this may be as simple as having a 'transfer size=0' on the experiments)

 

As far as open experimentals go, I believe the original plan was to fix the patcher so that KSP store users could also participate in the next round of open experimentals(presumably GOG users as well), but that was a stated goal at the beginning of the 1.1 open experimentals and plans may well have changed since then.

[Laie]

1 hour ago, Terwin said:

I would assume that you put in some sort of 'one way transfer' mechanic so that I cannot store a copy of my gravioli readings in a spent goo canister...(this may be as simple as having a 'transfer size=0' on the experiments)

It's even simpler than that. Science experiments ans science container are two distinct things.

I'm having difficulties putting this into words. But just use two Kerbals on EVA in order to take Soil Samples and give them to each other. You'll see what I mean. A Kerbal is both, two scientific experiments (take soil sample, take EVA report) as well as a science container.

[Nerfclasher]

I have a dream. I hope one day Squad will provide a configuration parameter with the max render/physics distance, that way at least it will be up to the user to decide a sensible value for his computer

[IncongruousGoat]

Does anyone know if the bug whereby service bays don't protect seated Kerbals within from aerodynamic heating has been resolved yet?

[Alshain]

8 hours ago, StahnAileron said:

Aw, crap. I love his mods. Even if I didn't have a real use for BDA, ALG was a godsend. His parts pack was useful as well. (The XJ-44 Vektor was wondrous, if heavy as hell.) Looks like only BDA was picked up for maintenance. I hope his other mods find new homes as well. (Or he comes back later.) Kinda wish ALG was stock, but it requires a custom plugin to work. And considering the changes to land gear in 1.1.x+, it would need a rewrite/overhaul. I don't foresee SQUAD integrating it. (ALG was perhaps the best "adjustable" anything available for KSP in terms of mechanics and functionality. It was pretty seamless in the latest version.)

I'm certain someone will pick up ALG, but until the wheels get fixed there is no point. 

[Alshain]

2 hours ago, Laie said:

It's even simpler than that. Science experiments ans science container are two distinct things.

I'm having difficulties putting this into words. But just use two Kerbals on EVA in order to take Soil Samples and give them to each other. You'll see what I mean. A Kerbal is both, two scientific experiments (take soil sample, take EVA report) as well as a science container.

Maybe this will help

Edited September 1, 2016 by Alshain

[Frank_G]

Sounds very good. I really hope, i have a good mission idea, once the new version is out  - but hey, the days get darker already. Space Adventure upcoming

[StahnAileron]

5 hours ago, Alshain said:

I'm certain someone will pick up ALG, but until the wheels get fixed there is no point. 

Well, yeah, I know about that. I was implying for that to happen post-1.2. (Or whenever wheels get truly un-FUBAR-ed.)

[Pthigrivi]

On 8/31/2016 at 8:25 AM, SQUAD said:

 http://i.imgur.com/U7bwR80.png

Really excited to play with all this  This is sort of a deeper question about science in general but I was wondering if the recovery value of any experiment needs to be less than 100%. One of the biggest concerns about career mode presently is science grind, and recovery rates less than 100% would seem to require either bringing redundant science parts or doing repeat missions to the same biome. Is this necessary? Another nice thing about 100% recovery rates would be that you could de-clutter the science UI because all you'd really need to see is one bar showing the transmit vs recovered value.

[RocketSquid]

Looks good! I've been hoping for something like these science return buffs for quite some time.

[KerbMav]

5 hours ago, Pthigrivi said:

Really excited to play with all this  This is sort of a deeper question about science in general but I was wondering if the recovery value of any experiment needs to be less than 100%. One of the biggest concerns about career mode presently is science grind, and recovery rates less than 100% would seem to require either bringing redundant science parts or doing repeat missions to the same biome. Is this necessary? Another nice thing about 100% recovery rates would be that you could de-clutter the science UI because all you'd really need to see is one bar showing the transmit vs recovered value.

First, you can of course edit some configs to get what you describe.

On the other hand you get almost all available science by transmitting first and running the experiment a second time to return it.

At the time the "loss" has any real numerical significance because of the science value getting high enough you most likely are raking in so much anyway it does not hurt either.

[Pthigrivi]

1 hour ago, KerbMav said:

First, you can of course edit some configs to get what you describe.

On the other hand you get almost all available science by transmitting first and running the experiment a second time to return it.

At the time the "loss" has any real numerical significance because of the science value getting high enough you most likely are raking in so much anyway it does not hurt either.

Yeah I guess what I mean is science collection is already pretty clicky, do all those extra steps really add anything? It might not seem like much but once you're going through half a dozen experiments and crew reports and eva's twice it really adds up. Especially now that we're talking about setting up satellite networks couldn't we make the tradeoff between recovery and transmission as simple and clear as possible? Im sure I could rejigger things on my end, this is really just a broader question about what makes for the best gameplay.

Edited September 2, 2016 by Pthigrivi

[Nerfclasher]

10 hours ago, StahnAileron said:

Well, yeah, I know about that. I was implying for that to happen post-1.2. (Or whenever wheels get truly un-FUBAR-ed.)

@StahnAileron @Papa_Joe has picked up all of bahamutod's mods and if you look through the the entire thread's op here you will find information there

[StahnAileron]

1 hour ago, Nerfclasher said:

@StahnAileron @Papa_Joe has picked up all of bahamutod's mods and if you look through the the entire thread's op here you will find information there

Okay... I have NO idea how I missed that mod list in the middle. Here I thought he only picked up BDA and not much else...

I noticed BD's engine pack isn't in the list. I already posted a question about it in that thread.

Thanks for the notice.

[Terwin]

13 hours ago, Pthigrivi said:

Really excited to play with all this  This is sort of a deeper question about science in general but I was wondering if the recovery value of any experiment needs to be less than 100%. One of the biggest concerns about career mode presently is science grind, and recovery rates less than 100% would seem to require either bringing redundant science parts or doing repeat missions to the same biome. Is this necessary? Another nice thing about 100% recovery rates would be that you could de-clutter the science UI because all you'd really need to see is one bar showing the transmit vs recovered value.

I see it as an incentive to bring along a scientist and a lab.

If you bring both of those, you can get all the science for each biome in one trip.

Kind of makes sense that if you want that last bit of science that you probably want to bring a scientist and lb with you, and anything less will mean lots of trips.

5 hours ago, Pthigrivi said:

Yeah I guess what I mean is science collection is already pretty clicky, do all those extra steps really add anything? It might not seem like much but once you're going through half a dozen experiments and crew reports and eva's twice it really adds up. Especially now that we're talking about setting up satellite networks couldn't we make the tradeoff between recovery and transmission as simple and clear as possible? Im sure I could rejigger things on my end, this is really just a broader question about what makes for the best gameplay.

The alternative to several clicks is for SQUAD to find some reliable, consistent, easily understood and unexploitable method to boost science returns based on who and what you have on your ship(and possibly including who collects the science), This would have the side-benefit of no longer knowing just how many copies of a given experiment you need to be able to get 100%, as it would be different based on which ship you were using(and it would need to keep track of the characteristics of the collecting ship at the time of the collection for each experiment, possibly including the local personnel/equipment during the extraction of the experiment).

Do I think SQUAD is capable of this?  Sure. 
Do I think it should be higher in the queue than other, more broadly appealing features?  nah.

And for those that  are really bothered by the current format, there are mods/config files.

[Merkov]

7 hours ago, Pthigrivi said:

Yeah I guess what I mean is science collection is already pretty clicky, do all those extra steps really add anything? It might not seem like much but once you're going through half a dozen experiments and crew reports and eva's twice it really adds up. Especially now that we're talking about setting up satellite networks couldn't we make the tradeoff between recovery and transmission as simple and clear as possible? Im sure I could rejigger things on my end, this is really just a broader question about what makes for the best gameplay.

Just my personal take on repeating science experiments, but part of me has always like the "lore" side of needing to do multiple materials bay, mystery goo, and surface samples to get full science value (there are other experiments, of course, but those ones always stick out to me). I guess I always think of it as silly from an "immersive" perspective to take a small rover to a biome, do all of your experiments, then leave and have all of the science that there is to have from said biome and never have to return again. Personally, I like have a little reward, tiny though it may be, for coming back. Maybe my first surface sample from the Mun's East Farside Crater had some small traces of some different material, and now Wernher Von Kerman wants more of that material for further study. I absolutely don't think it should be insanely grind-y (which is why I like that most of the available science can be retrieved the first time, as opposed to say only getting a quarter each time or something). I think the way it is now, if you just want to visit a biome, grab science, and go, the amount of leftover science shouldn't be so much that it will really hurt your program not to have it. If you do want to go back and get it, though, you get a little reward.

Having said that, I totally get why people install mods/modify configs/create MM patches to give full science for recovery. Different people want to play their game differently. I've just found that the stock way just kind of works for me.

(Basically, you are totally right. It is VERY clicky, but for some reasons I like clicky. Weird)

[Bombaatu]

I'd like to see science-over-time, much like ScanSAT does with altimetry & biome scans. I could see something like this:

Science experiment on rover or ship w/o Scientist on board: gain x% of max per (time period). If a Scientist is present (either as crew or within, say, a few hundred meters) OR if a Scientist is within short-range communications range of a rover/probe (say, within the range of a Communicatron16), then the science gain is x + (1/2x * Scientist level) % per (timePeriod). The experiment consumes EC while it is running & can be paused. It can also be stopped at any time & the current science transmitted.

Example: say the Materials Bay experiment had a base of 5% of max science gain (MSG) per 3 hours; it would require 60 in-game hours to get the full results. BUT, if a level 1 Scientist was present, then the gain would be 7.5% MSG/3 hours - 40 in-game hours. If a level 5 Scientist is available, this rises up to 17.5% MSG/ 3 hours, or just over 17 in-game hours for full results.

Perhaps I should put this into MOD ideas....

 

EDIT - nvm. It was pointed out to me that this is exactly with the MPL does. D'oh.

Edited September 2, 2016 by Bombaatu

[Pthigrivi]

5 hours ago, Terwin said:

I see it as an incentive to bring along a scientist and a lab.

Is it though? I think for mass reasons most people just do the experiment twice. I guess what I'm thinking is if this is what they're doing anyway why make them go through the extra step? It may seem like a small detail but grind and clickiness in the science system is one of the most often cited reasons people give for avoiding career mode altogether. I might be overlooking something but this seems like a pretty easy way to cut the problem in half and make the whole system much clearer in the process. 

Edited September 2, 2016 by Pthigrivi

[Terwin]

47 minutes ago, Pthigrivi said:

Is it though? I think for mass reasons most people just do the experiment twice. I guess what I'm thinking is if this is what they're doing anyway why make them go through the extra step? It may seem like a small detail but grind and clickiness in the science system is one of the most often cited reasons people give for avoiding career mode altogether. I might be overlooking something but this seems like a pretty easy way to cut the problem in half and make the whole system much clearer in the process. 

It works on me is all I can say.

Even on Ike and Duna, where I no longer have much need of additional science, I will usually biome-hop with my ISRU-equipped ship so as to get full science from several biomes just because of my completionist tendencies.

(And in 1.1 I take an extra of each experiment for loading into the lab for research on the way home/elsewhere)

 

Of course in 1.2 I may well just slap on a few extra probe-cores to hold the duplicate data instead(at least for the Kerbin system).

[KerbMav]

It really only is grindy when trying to get to the 100% from everywhere and not settle for one transmission and one return - plus maybe one instance for the lab. Also EVA and crew reports transmit fully and are therefore easy fodder for the lab.

If one would rather get the equipment one wants to use to go beyond Mun or Duna sooner, the science gain slider in the difficulty settings comes handy.

And clickyness is greatly reduced by binding all experiments to one action group.