[outdated] Caustic Arrow Solo Map MFer (20/300+)

Spoiler
"
Serleth wrote:


Quality is a function of stress testing. They are human. Certain humans will notice things other humans do not. By attritting the process, you're more likely to find issues.

I'm not really going to get further into the speculation of how well they do or don't test things because quite frankly, we just don't know.

I could go into this quite in depth - software quality has been my profession for 20+ years, however, I'll try to keep it short. ;)

Bugs have a natural distribution across severity and likelihood. All development teams tend to create about the same percentage of crash vs cosmetic, for example. Given that the bugs we face in POE are non-critical - the game doesn't crash, the servers don't crash, our player data isn't corrupted, etc, etc, I can emphatically say this game is rather well tested.

The issues that we *do* face are game balance, supporting skills that are triggered when they should not be, items dropping in places where they cannot be picked up, etc. These are not bugs that any product owner would (or should!) stop a deployment over.

Development teams don't attempt to ship a bug free product - they try to ship a product with the *right* bugs, the ones that don't impact the customers badly enough to affect satisfaction levels. Bugfixing past a certain point is cost prohibitive - even mission critical software that can kill people if it fails badly - air traffic control, NASA, etc - is only tested for double fault bugs at most.

Short... ish, anyway. :D
Current league IGN: Teldra_Anc_LAD
Feel free to message me in game if I'm on.
"
grimjack68 wrote:
Spoiler
"
Serleth wrote:


Quality is a function of stress testing. They are human. Certain humans will notice things other humans do not. By attritting the process, you're more likely to find issues.

I'm not really going to get further into the speculation of how well they do or don't test things because quite frankly, we just don't know.

I could go into this quite in depth - software quality has been my profession for 20+ years, however, I'll try to keep it short. ;)

Bugs have a natural distribution across severity and likelihood. All development teams tend to create about the same percentage of crash vs cosmetic, for example. Given that the bugs we face in POE are non-critical - the game doesn't crash, the servers don't crash, our player data isn't corrupted, etc, etc, I can emphatically say this game is rather well tested.

The issues that we *do* face are game balance, supporting skills that are triggered when they should not be, items dropping in places where they cannot be picked up, etc. These are not bugs that any product owner would (or should!) stop a deployment over.

Development teams don't attempt to ship a bug free product - they try to ship a product with the *right* bugs, the ones that don't impact the customers badly enough to affect satisfaction levels. Bugfixing past a certain point is cost prohibitive - even mission critical software that can kill people if it fails badly - air traffic control, NASA, etc - is only tested for double fault bugs at most.

Short... ish, anyway. :D


Interesting. Thanks for the insight.
Jul 27, 2011 - Sept 30, 2018.
My 2c on design/testing:

Spoiler
I've heard in several places that GGG developed a lot of their design philosophy from Mark Rosewater, head developer of Magic the Gathering... I used to read a lot of his articles back in the day... and play the game before I decided it was too much of a moneysink...

Magic the Gathering is basically designed around meta rotation... new expansions come out every 3 months... and in the new model rotate out in a year's time (used to be 2yrs)... ARPGs can't just rotate out skills/content in the same way, but they do comparable things with balance that MTG can't do to card balance directly...

MTG also has several testing errors in their past... mostly linked to deck control... Skullclamp back in Mirrodin gave a creature +0/-1 and allowed a card draw on death, which was OP since you could equip to a 1/1 creature/token and basically get cards for free on repeat... and a few years later with the Planeswalkers and Jace, Mind Sculptor... I remember hearing that card was so expensive ($100) you could buy packs/boxes of cards and sell that 1 card back and make money or cut even...


And I kinda hope HoWA stays... wouldn't mind playing it again... really enjoying ST... and I've read a few things secondhand that say people think devs have implied Breaches will stick around in some form. (convoluted, I know)
Prediction Time :-)

Make 3 predictions for next league.
Spoiler

Good stuff from Serleth, Grimjack (never knew that you were software QA) and Piros on my previous thread.

Just for fun, I'm going to make a few predictions for next league. What are your predictions?

1. Clear Speed Will Continue to be Extremely Important

Why?
Spoiler

I think GGG has nailed it this league. Certainly player participation is up.

GGG rightly changes things from league to league - perhaps unintentionally - searching for what's popular.

The clearspeed meta this league has been extremely popular - thus, I expect more of the same next league.

At the same time, development for the next league began way before the popularity of the current league was known. So probably mob-density will be shoe-horned into the next league somehow.

Moreover, I expect challenges to require even more time commitment next league than this league - why? Because GGG needs high player participation throughout the league - so the 36-challenge reward next league will involve at least one time-sync and that time-sync will be even more time consuming than Breach league.


2. Race Rewards Will Be Far Better

Why?
Spoiler

In spite of this league's popularity - player participation trends continue. High at league start... tailing off as the league continues... and dropping substantially in the 3rd month.

GGG will likely try to address this somehow... maybe race rewards will be better is my best guess.

Another possibility - shorter leagues. But, I don't think that this will happen - GGG doesn't have the resources to turn out 2-month leagues with high quality.


3. The Best Skill Next League Will Require a New (Probably League-Specific) Unique

Why?
Spoiler

Beta testing is broken and GGG doesn't seem to realize it. Virtually no one will beta test new items given how (I think) beta testing now works. As a result, one or more items will be far too powerful - but not adequately beta tested since it won't be obvious that the item(s) are so good and because you'll need to go out of your way to get some not-obviously good items during beta test.

In contrast to this - new skills, will be thoroughly vetted. It will likely be obvious that a new skill is OP and, as such, these skills will attract a bunch of beta testers. And even if not obvious, trying out a new skill has no barrier - equip the gem and go from there - as opposed to testing a new unique - you've first got to acquire the item, and then build around it.

If a new skill is actually OP, beta feedback will cause it to be nerfed - either immediately from testers... or from the community at large when new content sharing is approved.

That said, one or more new skills will likely be Tier 1 - otherwise, virtually no one will play them.

Last edited by hankinsohl on Feb 10, 2017, 1:05:03 AM
"
Piros wrote:

And I kinda hope HoWA stays... wouldn't mind playing it again... really enjoying ST... and I've read a few things secondhand that say people think devs have implied Breaches will stick around in some form. (convoluted, I know)

Me too :-).

Next league development began long ago - before the popularity of this league was known. So Breach inclusion, if it occurs, will be something simple as there won't be a lot of dev time to implement this.

My guess - Breaches will occur but with less frequency - basically the same treatment given to Essences.

With respect to HoWA - it's going to get the nerf bat - no way escaping this. Post-nerf will it still be worth pursing? Yes, I think so. But some other new item will almost certainly eclipse the nerfed HoWA.
"
hankinsohl wrote:
"
Piros wrote:

And I kinda hope HoWA stays... wouldn't mind playing it again... really enjoying ST... and I've read a few things secondhand that say people think devs have implied Breaches will stick around in some form. (convoluted, I know)

Me too :-).

Next league development began long ago - before the popularity of this league was known. So Breach inclusion, if it occurs, will be something simple as there won't be a lot of dev time to implement this.

My guess - Breaches will occur but with less frequency - basically the same treatment given to Essences.

With respect to HoWA - it's going to get the nerf bat - no way escaping this. Post-nerf will it still be worth pursing? Yes, I think so. But some other new item will almost certainly eclipse the nerfed HoWA.


They've already hinted at in the SF meetup posts on reddit that Breaches are likely to be either sextant or Zana mods, or something similar.

I think it'd be a mistake for it to not go core completely, but if it doesn't, that'll mean that HoWA probably won't be nerfed on the basis of its increased exclusivity. It'll end up being a chase item worth probably 10s of ex early on.

Hell, even if it does get the Essence treatment, I wouldn't expect an adjustment. Instead of multiple breaches per map, you'd be getting one breach every 5 maps, making Esh farming significantly more rare. In and of itself, that might be enough to warrant not touching it; since GGG is prone to adjusting drop rarity without notification, you might see that get bumped in order to keep the item even more chase without requiring any changes to the item itself.

*shrug*

Wait-and-see for me.

My only prediction is that Mathil will make 1c items worth half-exalts again :P
Jul 27, 2011 - Sept 30, 2018.
Last edited by Serleth on Feb 10, 2017, 2:45:12 AM
Grats on 40, Piros!

"
Serleth wrote:
"
hankinsohl wrote:
Please supply links to GGG statements about intentionally introducing/making available skills that are much worse than average - to me, this makes no sense whatsoever. But if GGG has taken this tack - well, whatever.


This was made a while ago, either in a video or over a year ago. I don't recall where it was.

Chris said it in the Lioneye's Watch Podcast, the episode with massive flask changes announced for 3.0. He said it not just about skills, but uniques and other things as well.

Hank, I don't think Blight is one of the bad skills right now. People are doing really nice things with it in PvM and it's an instant killer in PvP if you have a way to get close to the caster/LA/EA. It's just unpopular because it's not clear speed meta, and clear speed sucks anyway!
"
Chris said it in the Lioneye's Watch Podcast, the episode with massive flask changes announced for 3.0. He said it not just about skills, but uniques and other things as well.


Nice memory. Thanks for sourcing it. I thought it was less recent than that.
Jul 27, 2011 - Sept 30, 2018.
"
Serleth wrote:
"
Chris said it in the Lioneye's Watch Podcast, the episode with massive flask changes announced for 3.0. He said it not just about skills, but uniques and other things as well.


Nice memory. Thanks for sourcing it. I thought it was less recent than that.


The post explaining the design of Shrapnel Shot talked about how the skill was specifically designed for a specific level range of characters, and races.
"
ChronoExile wrote:
The post explaining the design of Shrapnel Shot talked about how the skill was specifically designed for a specific level range of characters, and races.


Ah yeah, that might've been what I was thinking of.
Jul 27, 2011 - Sept 30, 2018.

Report Forum Post

Report Account:

Report Type

Additional Info