Squadcast Summary 2015/07/03

[regex]

I think it's unlikely that the stock dV meter will have anything like KER's amount of data or customization, so I'll keep using KER for as long as it's developed.

I suppose there's that. It's not like Squad is known for providing detailed information to aid the player, after all...

[Joonatan1998]

Good luck running unmanned missions.

They are a lot easier than manned ones.

Much less delta-v required (unless you like unmanned return missions.), with less mass.

So I don't have I would have any problems.

Making manned missions easier doesn't make unmanned ones harder.

Edited July 4, 2015 by Joonatan1998
I forgot the possibility of unmanned return missions.

[Red Iron Crown]

They are a lot easier than manned ones.

Much less delta-v required, with less mass.

So I don't have I would have any problems.

Making manned missions easier doesn't make unmanned ones harder.

Er, delta-V requirements don't get lower by switching to an unmanned vessel. Delta-V is so useful a metric because it ignores vessel mass; a one ton probe and a 500t mothership require exactly the same delta-V to go to a particular destination.

[cantab]

I wouldn't want stock to go as far as KER in terms of detailed stage-by-stage breakdowns anyway. Delta-V calculations can be complicated and even KER gets it wrong sometimes. I recently ended up having to redo a lander design because KER had told me misleading figures, costing me a fair bit of time in the VAB, and it's a good thing I realised it then and not when I was trying to land! And then there are situations where the delta-V you get depends on the flight plan, such as Apollo-style missions, and therefore no computer program can ever give a definitive delta-V figure without being told the flight plan.

So I think I'd be content with something like first/current stage only. That should be reasonably safe against giving misleading figures, straightforward for Squad to program, and still very useful for all but the most complicated of designs.

Er, delta-V requirements don't get lower by switching to an unmanned vessel

What does make delta-V requirements lower is making a one-way mission rather than a return. Most players do one-way probes and return Kerballed missions, though there's no reason one can't do return probes and/or one-way Kerballed missions.

[Scotius]

I wonder when SQUAD will divulge first data about KSP performance in U 5? Will improvement be noticeable immediately, or is it going to be mostly "under the hood" thing?

[basic.syntax]

Delta-V should simply be displayed for the rocket, [...] I serious hope they rethink this decision.

I'm surprised you are surprised by this! You've been following the news all along, they've been speaking consistently. As I said in previous comment, I expect a close parallel to the implementation of pilot skills. The stock game paradigm is established.

Squadcast Summary 2015-03-20:

I am not 100% sure that we can put in the UI elements that we want to do for the Kerbal skills, so you can see delta-v for example

Reddit AmA for v1.0 launch:

The dV readouts were actually part of a much larger feature to expand on the skills for engineer and scientist kerbals.

Good luck running unmanned missions.

Wouldn't surprise me if they upgraded the probe cores, as they enable the various pilot skill abilities for unmanned craft. Edited July 4, 2015 by basic.syntax

[Joonatan1998]

Er, delta-V requirements don't get lower by switching to an unmanned vessel. Delta-V is so useful a metric because it ignores vessel mass; a one ton probe and a 500t mothership require exactly the same delta-V to go to a particular destination.

Manned missions usually are return missions, unmanned ones usually not.

Of course there are players who do permanent bases, "bases" or suicide missions, and players who do unmanned return missions, but the game really makes returning kerbals important (they are expensive).

[Red Iron Crown]

Fair point.

I may think of my kerbals as just a bit too expendable.

[regex]

I'm surprised you are surprised by this! You've been following the news all along, they've been speaking consistently.

I'm not surprised by it at all, I'm saying it's a really stupid design decision that should be changed.

[Eric S]

I wonder when SQUAD will divulge first data about KSP performance in U 5? Will improvement be noticeable immediately, or is it going to be mostly "under the hood" thing?

I'm looking forward to hearing about this as well, even if it's bad news just so that I can set reasonable expectations. Too much is changing to assume that the microbenchmarks we've seen will be an accurate representation.

For those that haven't been following, here's what I think will be the most significant factors in performance improvements:

1) PhysX optimizations: microbenchmarks indicate that for connected rigid bodies, the version of PhysX that U5 uses peaks at about 50% faster than the version that U4 uses.

2) Multithreaded physics simulation: I expect this to not affect single craft and only provide a benefit when there are multiple craft inside the physics bubble. Creating a general purpose physics model that can use multiple threads on a single set of connected rigid bodies isn't something that's been done yet to the best of my knowledge.

3) The new UI code: The consolidation of the existing three types of UI code, each with their own overhead, down into a single set of UI code should have a positive effect, though I'm not sure how noticeable it will be. Squad has commented that one of the types of UI code that's getting replaced was particularly bad in this regard, so it might actually be noticeable all on its own.

[Brotoro]

...I thought KSP players were generally intelligent people, perhaps I'm mistaken...

Being generally intelligent, I can't understand why astronauts would have to be calculating their ship's delta-V values when this should be a job for the engineers back at KSC. But perhaps I'm mistaken.

[FlowerChild]

This is will be the stupidest game design decision they've made. Delta-V should simply be displayed for the rocket, it's something that mission control and the rocket designer know from the beginning, plus having a skill progression for simple math is literally idiotic.

Whether or not a rocket designer would have that information available in the consensus reality, is ultimately a realism argument man, and as you yourself pointed out recently, those are often just wrappers for an attempt to make things easier (as this one is). Also, I can see why they'd tie it to skills so as not to force this information down the player's throat in the early game when it is least required.

Personally, I'm happy that they're tying it to the engineer skill as that will likely make the deltaV indicator all the easier to mod out entirely

[Wjolcz]

Wouldn't surprise me if they upgraded the probe cores, as they enable the various pilot skill abilities for unmanned craft.

If all of them got the full ability to calculate TWR, dV, etc. I would be actually cool with that.

[BudgetHedgehog]

Cheers for doing this, been a real headache at work recently. All looks like good stuff, excited to see the new shaders and steering modules.

[regex]

Whether or not a rocket designer would have that information available in the consensus reality, is ultimately a realism argument man

No man, it's pretty much a common sense argument.

those are often just wrappers for an attempt to make things easier (as this one is).

While I can accept it as an argument against making things easier, there is a big difference between an interesting game mechanic and utter stupidity that further reinforces the role of a single class of Kerbal. Seriously, why have classes at all? The early and mid-game will boil down to hiring two to three pilots and scientists to help level up your multitude of engineers who pretty much do everything in the late game.

[Wallygator]

Information restrictions based on "Kerbal skill".

Bah Humbug.

A complete waste.

It will force me to stay sandbox. Which isn't a bad thing considering the complete screw-up career is at the moment. Sad to see it continue. Career and skill both need a complete redux.

[FlowerChild]

No man, it's pretty much a common sense argument.

As you wish man. We've argued this subject before and while we are of similar opinions on many topics, are obviously of vastly different ones on this one, and it's one of the rare instances where I strongly agree with one of Squad's design decisions in not having a deltaV indicator in the first place. I've actually taken them sticking to their guns on that point as a sign of hope for a long time now, in not always caving to community pressure, but sometimes adhering to a design vision.

To me, the degree of information you present to the player is an integral part of a game's design. So, you say "common sense", I say "gameplay squashing", we go back and forth ad nauseum and eventually give up trying. May as well just cut to the chase

It should be noted that in these discussions of ours I haven't really been arguing against the inclusion of a deltaV indicator overall. Just some of the individual points you've been making for them, which I don't think are particularly strong, the realism argument and "two sides of the same equation so you may as well rip out the existing deltaV for maneuvers indicator" included.

I do think that a valid debate could be had over this on the degree of information that the player should have access to and what makes for a more fun experience overall.

While I can accept it as an argument against making things easier, there is a big difference between an interesting game mechanic and utter stupidity that further reinforces the role of a single class of Kerbal. Seriously, why have classes at all? The early and mid-game will boil down to hiring two to three pilots and scientists to help level up your multitude of engineers who pretty much do everything in the late game.

Yeah, I do agree that the class mechanic is weak at present and doesn't bring much to the table. I think it *can* become something interesting with time but will require a lot more work, and probably entirely new gameplay mechanisms (like the engineer class makes a lot more sense in the context of something like KAS...but not stock).

Ultimately, I do like the idea of incentivizing multi-man missions through specialized roles, but I really don't think that's where stock is at, at present. IMO, there's some good to the base idea, but the current execution is lacking. I also think that without some kind of downside for multi-man missions (life support requirements or such), it all winds up being fairly meaningless as you could just stuff a few more clowns into the car anyways.

Edited July 4, 2015 by FlowerChild

[basic.syntax]

The early and mid-game will boil down to hiring two to three pilots and scientists to help level up your multitude of engineers who pretty much do everything in the late game.

While I agree with this statement to a degree - Squad isn't done yet, additional features for scientists hinted below, may help fill in the "this kerbal class is useless" syndrome you are feeling today. (I would argue the new MPL makes scientists very useful in the mid-game, and a supplemental income stream in late-game, when players switch strategy Science >> Funds)

The dV readouts were actually part of a much larger feature to expand on the skills for engineer and scientist kerbals.

Edit:

Information restrictions based on "Kerbal skill". Bah Humbug.

My prediction: watch for engineer skill features to make an appearance in probe cores, just like they did for piloting. That's also a way we could get dV / TWR in the VAB... based on the level (type) of probe core you have unlocked. ( Or a new "engineering computer" part. Or info based on VAB upgrade level. If I make enough guesses, one of them might come true...) Edited July 5, 2015 by basic.syntax

[Brotoro]

q: will the gimbal be nerfed? it's way too fast moving also the sas always overshoots the nodes

a: "we might go back into sas and gimbal speed in the future. it's not in the short term plans but is part of the gameplay polishes we have taken a peek at.

right now they can react too quickly... more than the sas wanted them to move and then has to compensate an it's not hyper-efficient. It's not top priority but it's definitely something we can look into"

I'm sad to hear that. The SAS pointing problem is really bad, and if I were Squad I'd fix such an embarrassment right away.

[Zucal]

The early and mid-game will boil down to hiring two to three pilots and scientists to help level up your multitude of engineers who pretty much do everything in the late game.

Y'know, what might be a lot better than three classes of Kerbals is a kind of tech/skill tree for each Kerbal. You can spends increasing amounts of funds to allow Kerbals to specialize, or do everything. A completely untrained Kerbal might be like a current 0-star one, but a highly-trained Kerbal could repair solar panels and radiators, etc.That way, a Kerbal has a bit more individuality and value- that along with badges for where they've been would vastly improve that chunk of the game.

[foxkill2342]

I'm sad to hear that. The SAS pointing problem is really bad, and if I were Squad I'd fix such an embarrassment right away.

This would require someone from the team to actually play the game outside of the squad cast.

[Rocket Farmer]

Per all the hate for them not including KER. You guys do know it's a simple math calculation to figure out your DV. Just have excel running in the background and plug in your total mass and total fuel (both of which are listed now).

Personally I play totally stock and have no issues. Infact using Excel greatly reduces my build sizes/time building/ and failure rates as I can quickly model each stage and determine where I can reduce mass/fuel, try different engine combinations quickly, ensure my TWR is suitable in any given stage, and still accomplish my mission.

Frankly I don't know how you guys get anywhere with the little bit of data you get from KER.

[Superfluous J]

Per all the hate for them not including KER. You guys do know it's a simple math calculation to figure out your DV. Just have excel running in the background and plug in your total mass and total fuel (both of which are listed now).

But I'm not an engineer. Even though I've flown to other planets and back I can't even change a tire, little alone calculate simple math.

And regarding using a spreadsheet program to play a video game... No. Just no. I won't do it. This is not Kerbal Accountancy Program or Kerbal Charts And Graphs Program. I want the stuff in the game in an easily consumable way so it stays fun and doesn't become stale. I deal with spreadsheets at work. I don't want to futz with them during my limited time playing a game.

And I've documented on video for the past 2 years (almost) how far I can go with KER. Spoiler Alert: Everywhere.

[FlowerChild]

And I've documented on video for the past 2 years (almost) how far I can go with KER. Spoiler Alert: Everywhere.

And I can go everywhere without it...and without a spreadsheet...and without performing any calculations at all other than stuff like energy or life support consumed over time vs total carried.

That's not an argument against a deltaV display being integrated into stock. At this point I don't think I really care either way. It is an argument however against the assumption that it's in any way necessary in order to play the game and do everything in it.

It's really not. For reference, I'll link my first ever post to these forums where I specifically stated I didn't want anything that provided additional information or gameplay aids:

http://forum.kerbalspaceprogram.com/threads/40162-Looking-to-start-installing-mods-to-provide-a-*deeper*-gameplay-experience

Didn't stop people from trying to ram the "need" for them down my throat however

I stuck to that in these past two years with a couple of minor exceptions like PreciseNode and Waypoint Manager that I feel make up for deficiencies in the stock interface. Interface nuisances aside I personally find stock KSP to be extremely easy to play, and that was true when I started playing as well. The last thing I've ever wanted were tools to make it even easier like a deltaV readout.

Edited July 5, 2015 by FlowerChild

[Superfluous J]

I was specifically addressing:

Frankly I don't know how you guys get anywhere with the little bit of data you get from KER.