These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE Information Portal

 
  • Topic is locked indefinitely.
 

New dev blog: Starbase tweaks: an update

First post First post
Author
Azurun Li
Number One Lucky Golden Dragon Buffet
#61 - 2011-11-23 16:50:09 UTC
The changeover plan is a good one. I support leaving it as-is.
Lake
The Praxis Initiative
Gentlemen's Agreement
#62 - 2011-11-23 16:54:23 UTC
CynoNet Two wrote:
CCP Greyscale wrote:

If you want to argue that "no fuel blocks at all" is better than "fuel blocks without a handover script", then that's an interesting conversation (although only in a theoretical sense this late in the day), but "fuel blocks and a handover script" just isn't something we could justify considering for this release. We only have so many developer hours to work with in a given expansion so everything is a zero-sum decision - adding time to one thing means taking it away from something else.


I feel like Captain Obvious pointing this out, but...

1) Go ahead with the November 29th as-is. No need to mess with code, no feature changes.
2) Let players build up supplies of fuel blocks from the newly seeded BPO's.
3) Cancel the fuel change-over 2 weeks after Crucible.
4) Take your time writing and testing a switch-over script. It's not exactly a complex task, just a case of iterating through each POS in the database and replacing X fuel with Y blocks.
5) Test it to destruction with QA time after the main patch is done and dusted. Deploy in the New Year.

Everyone now has nice stocks of the new fuel ready, doesn't have to rush around just before the holidays manically saving towers, and is generally happier.

You seem you determined to push out the entire change ASAP at the expense of forcing thousands of players to expend many hours in unecessary work. This isn't a trade-off between QA time versus other features - it's a trade-off between putting out a new feature RIGHT NOW, or pushing out a new feature properly in a manner that accomodates players.
There is little benefit in pushing this change in a sea of chaos before Xmas over doing it in a sea of calm after Xmas.

Take your time.


Typically the best insight is obvious in hindsight (iow, good post). That distinction that the patch day can come and go on schedule while the DB script can be written and tested later is important.

The expected argument here is that the developer time after patch day is itself already schedule for other tasks, however I think folks would be content to give up a day's work on the patch after this one in exchange for this changeover being less of a headache. Even if it means the changeover happens three, four, or six weeks after patch day instead of the planned two.
Kismeteer
Bat Country
Pandemic Horde
#63 - 2011-11-23 16:55:41 UTC  |  Edited by: Kismeteer
If you wanted to make the task a little easier on your DBAs, you could also just replace by m^3 full, which is an easy measurement. Because of the huge Heavy Water and Liquid Ozone increases, a 1-to-1 mapping might be difficult.

Converting a pos of 50k m^3 of fuel to 10k fuel blocks of the correct version is fine, though I'm sure some people will try to game the system by putting up towers of 110k of the lowest cost pos fuel. It's pretty easy to see when a pos is trying to be exploited if you wanted to stop it.

And if anyone at :ccp: says 'But it's too many POS!' think about the balancing act all those pilots will have to do with a very large commodity. To you, it's a couple database record and SQL statements. To us, it's several hours of work, per pos.
szaiboT
Operation Myriad
Galactic Greybeards
#64 - 2011-11-23 16:57:37 UTC
Just want to add my part of story which is probably true for most people running POSes in WH space

This does NOT simplify anything. It complicates things for me.

Firstly I believe most people in WH space do use their planets for PI to cut costs on running POS and to CUT the amount they need to howl.

With this "feature", I either have to add additional step of producing those cubes

or simply howl from known space, and therefore instead of howling just ice products I have to howl everything.
Kismeteer
Bat Country
Pandemic Horde
#65 - 2011-11-23 17:01:29 UTC
szaiboT wrote:

or simply howl from known space, and therefore instead of howling just ice products I have to howl everything.


Word you're looking for is haul. But yes, this is more work. Plus, you have to haul extra HW and LO3 volumes too.
Jack Dant
The Gentlemen of Low Moral Fibre
#66 - 2011-11-23 17:01:34 UTC
szaiboT wrote:
Just want to add my part of story which is probably true for most people running POSes in WH space

This does NOT simplify anything. It complicates things for me.

Firstly I believe most people in WH space do use their planets for PI to cut costs on running POS and to CUT the amount they need to howl.

With this "feature", I either have to add additional step of producing those cubes

or simply howl from known space, and therefore instead of howling just ice products I have to howl everything.

You must have a really sore throat after all that howling.

What happens in lowsec, stays in lowsec, lowering the barrier to entry to lowsec PVP: https://forums.eveonline.com/default.aspx?g=posts&m=476644&#post476644

Ingvar Angst
Nasty Pope Holding Corp
#67 - 2011-11-23 17:04:55 UTC
Kismeteer wrote:
szaiboT wrote:

or simply howl from known space, and therefore instead of howling just ice products I have to howl everything.


Word you're looking for is haul. But yes, this is more work. Plus, you have to haul extra HW and LO3 volumes too.


Coming soon to theaters near you...

Twilight: Wormhole.

Six months in the hole... it changes a man.

CCP Greyscale
C C P
C C P Alliance
#68 - 2011-11-23 17:09:35 UTC
All that pushing the testing* into the future achieves is to add more uncertainty and risk, unfortunately. We'd still need to spend a significant amount of testing time on making sure the script works, we'd still have to prioritize testing the upgrade over testing other stuff, which would still mean cutting a bunch of features to get this in - and it still probably wouldn't be worth it. It's not a case of "having time available", because we had "time available" for this expansion - it's a case of "being worth spending the time on this rather than something else".

*Anyone saying that writing the script should be trivial is, uh, not speaking from a position of authority, but it's the testing time that's the real workload, because every time we want to test if it works (which we need to do at least once and potentially two or three times to test fixes for problems caught in earlier runs) we have to actually run the upgrade on a copy of TQ, which takes a large amount of setup and prep time. (And before anyone says "surely that's really simple", it's not, and that's really all there is to say about it :P)
Entity
X-Factor Industries
Synthetic Existence
#69 - 2011-11-23 17:12:17 UTC
Victor Valka wrote:
Thanks for removing some unnecessary complexity, CCP! Big smile


The complexity is still there. It's just dumped on the guy making the cubes now.

╦......║...╔╗.║.║.╔╗.╦║.╔╗╔╦╗╔╗

║.╔╗╔╗╔╣.╔╗╠..╠ ╠╗╠╝.║╠ ╠╝║║║╚╗

╩═╚╝║.╚╝.╚╝║..╚╝║║╚╝.╩╚╝╚╝║.║╚╝

Got Item?

CynoNet Two
GSF Logistics and Posting Reserves
Goonswarm Federation
#70 - 2011-11-23 17:13:58 UTC
CCP Greyscale wrote:
we'd still have to prioritize testing the upgrade over testing other stuff, which would still mean cutting a bunch of features to get this in


Why, what features do you have planned for January?
Trimutius III
Foundation 053
#71 - 2011-11-23 17:18:24 UTC  |  Edited by: Trimutius III
Sorry found an answer... Where was I looking?
CCP Greyscale
C C P
C C P Alliance
#72 - 2011-11-23 17:23:06 UTC
CynoNet Two wrote:
CCP Greyscale wrote:
we'd still have to prioritize testing the upgrade over testing other stuff, which would still mean cutting a bunch of features to get this in


Why, what features do you have planned for January?


We're always working on something
Lake
The Praxis Initiative
Gentlemen's Agreement
#73 - 2011-11-23 17:28:26 UTC
And the contention here is that we're willing to give up "something in january" for this, given that it won't push the release date of the other features in Crucible to get this feature in Crucible deployed in an elegant way.
Mnengli Noiliffe
Doomheim
#74 - 2011-11-23 17:40:33 UTC
to be honest, 5 minutes instead of 10 is simply no change.

I mean what is the point in introducing the need for manufacturing in the first place? why before you did not have to manufacture anything to fuel a starbase, and now you have? what kjnd of nerf is that?

If this is supposed to be a usability upgrade (instead of a usability nerf as it is now), just make this conversion free, or take some symbolic time, like 1 minute for 400 cubes.

With what it is now, we'll have all the high sec manufacturing facilities taken by only the fuel cubes production for years to come.

You might just say so if you wanted to nerf high sec manufacturing, not disguise it as a 'boost' to starbase management.
3rdTimeLucky
Perkone
Caldari State
#75 - 2011-11-23 17:46:56 UTC
I believe the number worked out earlier was between 3% and 5% of Empire manufacturing slots, which probably isn't "all the high sec manufacturing slots"; correct me if I'm wrong.

Plus this opens up another niche manufacturing area. I might even think about putting up a tower in high sec to produce some fuel blocks, hmmm.
Destination SkillQueue
Doomheim
#76 - 2011-11-23 18:10:41 UTC
Mnengli Noiliffe wrote:
to be honest, 5 minutes instead of 10 is simply no change.

I mean what is the point in introducing the need for manufacturing in the first place? why before you did not have to manufacture anything to fuel a starbase, and now you have? what kjnd of nerf is that?

If this is supposed to be a usability upgrade (instead of a usability nerf as it is now), just make this conversion free, or take some symbolic time, like 1 minute for 400 cubes.

With what it is now, we'll have all the high sec manufacturing facilities taken by only the fuel cubes production for years to come.

You might just say so if you wanted to nerf high sec manufacturing, not disguise it as a 'boost' to starbase management.


Cutting the manufacturing time in half is a large change. This feature is a usability improvement, but it also offers industrialists something new to do. It's sort of like why have guns use ammo when you could just have all the basic minerals in the cargo hold and guns would automaticly convert them to pew-pew. IT's not strictly necessary to have ammo, but it makes sense, is a lot more convenient than calculating how much of what minerals you need to carry for your ammo needs and it offers industrialist a new thing to produce and a new product to make profit from.

The effect on manufacturing facilities won't be as big as you think, because POSs will also produce them and there is no more fuel savings for not using up all your CPU and powergrid on a POS. This means people who didn't have their POSs fully utilized are now looking to make use of that extra capacity in some way, since they are going to be paying for it anyway. One good use for it is going to be a facilty, that is capable of producing fuel blocks for their own consumption and for sale to the market.
Ethilia
Freelance Excavation and Resistance
#77 - 2011-11-23 18:26:53 UTC
Since you have 6 days left (which is more than enough time for such an awesome team), can you make it possible to (re)name the corporate hangars and other PoS arrays?

Also, making fuel blocks 0.5m3 would be even more awesome for us poor logistics managers.
Knug LiDi
The Dark Space Initiative
Scary Wormhole People
#78 - 2011-11-23 18:42:47 UTC
Scrapyard Bob wrote:
Knug LiDi wrote:

The changes to the cost to produce reactions at POSs are very sensitive to operating costs, as our percentages are slim. I speak as someone who doesn't get the moon goo for free.


(Rehash of what I wrote up earlier in another thread.)

The only towers at a risk of costing slightly more are large towers. The cost savings on small/medium towers far outstrip the amount of ISK added back to the 30-day fuel costs by increased HW/LOz needs.

For the Amarr Large Tower, they would have saved 24M per 30 days under the new fuel pellet ingredient list. If you were not using all of your CPU before this, you would still save about 8M/mo. If you were not using all of your PG, then you'll end up paying about 21M more per 30d.


So, confirming those of us with large towers running with low PG/CPU usages will pay more for fuel.

I am glad with your analysis that the overall fuel cost impact will be relatively minimal - I had not hard calculated the actual impact yet - but in any event, we will be paying more at our large low-intensity POSs, but perhaps not as much as I had feared.



If only we could fall into a woman's arms

without falling into her hands

Hans Arienth
Hanson Heavy Industries
#79 - 2011-11-23 18:52:34 UTC
I, also, support the current transition plan. Bring on the jello fuel cubes!!!
ZaBob
The Scope
Gallente Federation
#80 - 2011-11-23 18:54:49 UTC
CCP Greyscale wrote:
All that pushing the testing* into the future achieves is to add more uncertainty and risk, unfortunately. We'd still need to spend a significant amount of testing time on making sure the script works, we'd still have to prioritize testing the upgrade over testing other stuff, which would still mean cutting a bunch of features to get this in - and it still probably wouldn't be worth it. It's not a case of "having time available", because we had "time available" for this expansion - it's a case of "being worth spending the time on this rather than something else".

*Anyone saying that writing the script should be trivial is, uh, not speaking from a position of authority, but it's the testing time that's the real workload, because every time we want to test if it works (which we need to do at least once and potentially two or three times to test fixes for problems caught in earlier runs) we have to actually run the upgrade on a copy of TQ, which takes a large amount of setup and prep time. (And before anyone says "surely that's really simple", it's not, and that's really all there is to say about it :P)


Of course the QA time is more than the script. I could write the script in 10 minutes or so; I can't imagine how long I'd want to estimate for the QA. With you 100% on that.

But -- without it, I think this entire FEATURE is not worth it. Or, at the very least, it will be a LONG time before the extra hassle is made up by any savings in subsequent hassle.

And without a volume adjustment, I think it is a serious PITA for WH operators. I brought this up before; others have brought it up, too. That, I think, you could take care of in a single stroke; you'd spend more time discussing the exact value than implementing and testing.

I think you've fallen in love with both the feature and the schedule. I think you need to back off and not rush this.

As it is, I would MUCH rather just not have it, and continue with the status quo. If you want to make my life easier, just reduce the volume on the ice fuels. Save yourselves a bunch of QA work.

I spend FAR more time hauling fuel, than I do calculating how much of each fuel. The complexity really doesn't bother me. The volume does. The volume of these blocks is FAR higher than the volume of ice fuels alone, which is all I current have to haul.

Can we get you, CCP Omen, and a bunch of other devs to go live in a WH for a couple months? It would be both fun for you, and educational.