This week our dedicated Evocati testers from the Community tackled Alpha 3.0 and allowed us to see performance and network issues we weren’t able to with our internal test group. This feedback gave us good insight on areas we can optimize to ensure that when 3.0 goes Live, it is a smooth, fun experience. As of Thursday, we were at 23 must fix issues to polish our shopping and cargo features, as well as working together with Evocati to process and fix bugs so we can get the build ready for the PTU and then go Live.
"This week our dedicated Evocati testers from the Community tackled Alpha 3.0 and allowed us to see performance and network issues we weren’t able to with our internal test group. This feedback gave us good insight on areas we can optimize to ensure that when 3.0 goes Live, it is a smooth, fun experience. As of Thursday, we were at 23 must fix issues to polish our shopping and cargo features, as well as working together with Evocati to process and fix bugs so we can get the build ready for the PTU and then go Live."
Burn Down, current status (Weekly Update):
The number of issues lies in 23, mind these are the list of bugs to deal with before pass update on the Evocati build that should focus on fixing/polishing the Shopping & Cargo Features.
Completed this past week:
Comms System UI.
Resuming last week: - 1 Completed Tasks - A total of 5 features left.
Bug-fixing changelog for the past week in the full schedule page and in Evocati patch notes of the pretty much daily builds.
"This week, we generated more than fifteen internal builds which allowed QA to confirm fixes and identify new critical bugs. Three of these builds were released to the Evocati for intensive testing. One persistent bug was a memory leak issue where the game (either on the client or server side) soaked up more and more system RAM until it crashed. The team has been squashing this bug whenever they find it, but with performance issues like this, there is rarely a single solution. As more mechanics and code are implemented, any one of them can cause the issue to return. The programmers have been working with the engineering team to develop new additions to the Debug system which will help identifying these issues sooner. There was also a consistent Serialized Variable crash that occurred when there were data mismatches in the game code between the client and server which caused a disconnection. The debugging work the engineering team is doing for the memory leak should also help resolve this issue as well.
The team’s been focusing on the comprehensive Shopping experience and resolving any issues related to shopping, kiosks or the economy. They’ve fixed a series of issues that ranged from spawning the player inside their bed (literally) to one bug that took out all of the items from all of the shops. "
Burn Down, current status (Weekly Update):
What's still in progress:
Continuing setup of the repair mission
Setting up basic steal mission
Implementing mission relevant dialogue to missions
Working on ship weapon pricing
Setting up the Shopkeepers in all the shops in game – working through issues with Subsumption
Testing Obj Container lighting exports – running into issues though where the lighting states aren’t working
Prepping the Mustang implementation file for the rework.
Continued tweaking Lighting, RTT, Open Physics Proxy Pass on all ships.
Continued wokring on the polish pass for Area18 props.
Continued converting props into new Usable system.
Continued working on Serialised Variables optimisations, and has been looking at Code 7 disconnect bugs.
The team has also been looking at potential optimizations and started performance R&D.
Continued working on Vid Comms.
Continued working on interactions.
Enabling airlock doors to open automatically when approached
Iterating on the Conversation system for characters
Working on HUD “Tidy Up” (e.g. – with no kit, helmet etc)
QATR is in now for headless clients (this system allows a developer to populate a server with AI Clients to test population-triggered bugs). First limit will be 24 headless, the next will be 100, if the first is successful.
Pawel has been doing some work on persistent spawning, has been bug fixing, and will move back to his proxy issues.
Bug-fixing changelog for the past week in the full schedule page and in Evocati patch notes of the pretty much daily builds.
All sounds nice and everything, but until its all in a working game, then its really not all that relevant, and what i mean by a working game, is that i don't mean parts of a game that work, but hats off to those who have some amazing levels of patience and have stuck with it this long, its almost the end of the year and although i am no expert, it doesn't sound like they are really any closer to completing this game than they were at the beginning of the year, i just hope that those people who have bought the game are not going to end up being disappointed, i just know that for me i don't have the kind of 'patience' that is needed when it comes to waiting for things i have paid for to be 'delivered', maybe 2018 will bring the game closer to fruition, but i have to wonder if it this time next year if we'll just be looking at more lists of updates.
They keep the numbers the same, but when completing some feature takes too long, they just move some of the planned features to a higher version number.
If you do this gradually enough , like over a year, little by little moving over features to future versions, it almost seems as if they stay close on schedule In version numbers anyway, not in actual planned features. 'Yeah, I am sorry, we have to move this feature and this feature over to next year, but don't worry, 3.1 is on schedule!'
Anyway, you could almost compare it to this company that keeps saying that there is enough money to finish SQ42, and with the box sales of SQ42 they would have enough to finish development of SC. Meanwhile they keep introducing more ridiculously priced ships that you can 'fund' the game with. Oh wait , that is the same company, doh.
I wonder if they plan to continue selling these packages after game release too. I would be in awe if they manage to do that. It would be like GTA Online, but then greatly improved. Shark cards, but then with a better pitch.
All sounds nice and everything, but until its all in a working game, then its really not all that relevant, and what i mean by a working game, is that i don't mean parts of a game that work, but hats off to those who have some amazing levels of patience and have stuck with it this long, its almost the end of the year and although i am no expert, it doesn't sound like they are really any closer to completing this game than they were at the beginning of the year, i just hope that those people who have bought the game are not going to end up being disappointed, i just know that for me i don't have the kind of 'patience' that is needed when it comes to waiting for things i have paid for to be 'delivered', maybe 2018 will bring the game closer to fruition, but i have to wonder if it this time next year if we'll just be looking at more lists of updates.
We will always be looking at lists of updates, as one releases it all starts to expect and wait the next one, I don't think this would be any different after the game is released.
The release and completion of the game in the case of SC are not tied together either, so to be seen when they decide to call that.
This week, the team generated another fifteen builds for the developers and QA to review, while three more made their way to the Evocati testers. One of the major victories this week was the first implementation of our new WAF build system, which will cut down the required time to generate a build from four hours to just one. When perfected, this will allow us to have more development and testing time between each build.
The team continued to focus on perfecting the Shopping experience this week, finally fixing a bug that prevented players from equipping items that they bought. Our daily review sessions focused on the physical stores in locations like Port Olisar, Grim HEX, and Levski, so we’re in the process of prioritizing those bugs in our overall fix-list. The rest of the team continued to tackle the ongoing bugs tied to commodity shopping that we discussed last week. One of the setbacks we encountered this week was a pair of Disconnection errors that would kick players from the servers. These bugs were discovered thanks to improvements made to increase server stability. As players could play for longer, they were now able to trigger these Disconnection errors. Finally, our Network team checked in a fix that will increase performance in the overall game experience.
We are continuing to improve the production schedule report to accommodate your feedback and next week we will have a revised format that highlights the remaining features that need completing before we go to a wider PTU and then live.
The team’s been focusing on the comprehensive Shopping experience and resolving any issues related to shopping, kiosks or the economy. They’ve fixed a series of issues that ranged from spawning the player inside their bed (literally) to one bug that took out all of the items from all of the shops.
Burn Down, current status (Weekly Update):
So the schedule is going to change format next week, I guess mostly because the Evocati is working with multiple update passes to fix and polish up part by part of the build.
(currently shopping, cargo and trading, next the new mission system)
But speculating this out I would say Citcon card will be SQ42, 3.0 has been already last month in GC.
SQ42? That thing they JUST missed being able to show off last year by DAYS. And then was a no show for 12 months. That SQ42?
Would be about f'ing bloody time if they did...
youll NEVER see ANYTHING that is PLAYABLE at any of these conventions. At least not playable on any server other than their own.
But SQ 42 should tell everyone something.
They WILL make a MAJOR announcement about SOMETHING and then next year THAT WONT be delivered either.
The fact that people fall for the same shell game year in and year out is amazing.
You can go back over the past 3 or 4 conventions (pick any of them) there isnt a single thing that highlighted any of them that is actually in the PTU, that includes stuff from 3 and 4 years ago.
Coincidence? Not hardly. But the bait and switch has gone on for so long people cant even remember what has been showcased and hyped and what hasnt.
Procedural cities anyone?
I wasnt going to bump this because predicting what these guys are going to do is about as hard as saying the sun will come up tomorrow (and it still comes up even if its cloudy).
Also the first line is pretty awesome since apparently they let peopel play 3.0 on THEIR servers and of course it ran flawlessly, yet the leaks from the avocados say 8-15 frame rates as long as you have 32 or more gig of Ram and the best vid card money can buy.
Also the first line is pretty awesome since apparently they let peopel play 3.0 on THEIR servers and of course it ran flawlessly, yet the leaks from the avocados say 8-15 frame rates as long as you have 32 or more gig of Ram and the best vid card money can buy.
That is what the leaks say:
"Q: How's performance in the most recent patches?
A: E and F were dogshit with constant 30K errors and CTDs. Patch G
gave me 60FPS on the US server for the first time, and it didn't drop me
to less than 30 FPS on a full server. "
Also the first line is pretty awesome since apparently they let peopel play 3.0 on THEIR servers and of course it ran flawlessly, yet the leaks from the avocados say 8-15 frame rates as long as you have 32 or more gig of Ram and the best vid card money can buy.
That is what the leaks say:
"Q: How's performance in the most recent patches?
A: E and F were dogshit with constant 30K errors and CTDs. Patch G
gave me 60FPS on the US server for the first time, and it didn't drop me
to less than 30 FPS on a full server. "
Have fun
I dont believe it. You dont go from constant crashes and errors to clear 60 FPS. There arent a lot of current games where you can get 60 FPS at top settings and a lot of rendering to do in high traffic areas. 30 top end MIGHT be semi believable with dips to 10-15 but even then not too sure.
But always nice to see what parts of guys breaking an NDA people will cherry pick and when they will cherry pick them. When the leaks were negative they were 'fake' when they put a nice spin on it theyre being quoted as 'proof'.
In the performance matter so what it was, is that they do have done server improvements and they have tested out this stuff in one recent build resulting in higher FPS performance and stress tested the servers, the thing is it ended up causing a lot of crashes so this is continuing development to work out the stability of that before it's properly added.
That's nothing but the point of ETF itself, there's a very flippant nature of this factor as testing endures, hoping they work out the stability of the changes they pushed before to maintain the network performance as they continue work.
In the performance matter so what it was, is that they do have done server improvements and they have tested out this stuff in one recent build and stress tested the servers, the thing is it ended up causing a lot of client and server crashes so this is continuing development to work out the stability of that before it's properly added.
That's nothing but the point of ETF itself, there's a very flippant nature of this factor as testing endures, hoping they work out the stability of the changes they pushed before to maintain the network performance as they continue work.
no idea what you just said and I cant even decipher what you think youre saying.
Bottom line is they have these movies they call gameplay then they have (alleged) people 'playing' the latest build, all running nearly 100% flawlessly. When not even the staunchest white knight can claim what they offer on the test bed runs even adequately and thats the stuff thats been there for months.
You cant watch a guy that is streaming (one of the half dozen) where you will find a stream that lasts more than 30-45 minutes before a CTD or a freeze up. Unless of course its the guy standing in a hanger facing a blank wall and him pontificating about the 'future' of SC.
The 'regular' guys know where ot go and what to avoid so their crashes have been minimized but even they fall victim to them. But get a random guy trying it out for the first time or the first time in awhile and it will be constant meltdown and freezes.
Thats why I always look forward to the dog and pony show when they do these demonstrations at the conventions because its so far from reality it could probably be considered 'fake' on that aspect alone. But since it goes way beyond that in terms of 'production' value its obviously faked in more ways than that. Or the way it runs is proof that its fake. People can draw their own conclusions.
In the performance matter so what it was, is that they do have done server improvements and they have tested out this stuff in one recent build and stress tested the servers, the thing is it ended up causing a lot of client and server crashes so this is continuing development to work out the stability of that before it's properly added.
That's nothing but the point of ETF itself, there's a very flippant nature of this factor as testing endures, hoping they work out the stability of the changes they pushed before to maintain the network performance as they continue work.
no idea what you just said and I cant even decipher what you think youre saying.
Bottom line is they have these movies they call gameplay then they have (alleged) people 'playing' the latest build,
Then go work on basic English comprehension before writing thesis on the forums and criticize others.
In the performance matter so what it was, is that they do have done server improvements and they have tested out this stuff in one recent build and stress tested the servers, the thing is it ended up causing a lot of client and server crashes so this is continuing development to work out the stability of that before it's properly added.
That's nothing but the point of ETF itself, there's a very flippant nature of this factor as testing endures, hoping they work out the stability of the changes they pushed before to maintain the network performance as they continue work.
no idea what you just said and I cant even decipher what you think youre saying.
Bottom line is they have these movies they call gameplay then they have (alleged) people 'playing' the latest build,
Then go work on basic English comprehension before writing thesis on the forums and criticize others.
I can neither confirm nor deny that those claims are correct or fake.
I have a hard time believing that someone with your dedication to this game is not a part of the testing group. You post things within an hour of them going on the official forum and you post here continuously throughout the day.
Why are you not a tester? Dont you want the game to be the best it can be?
"Sean (Murray) saying MP will be in the game is not remotely close to evidence that at the point of purchase people thought there was MP in the game." - SEANMCAD
I can neither confirm nor deny that those claims are correct or fake.
I have a hard time believing that someone with your dedication to this game is not a part of the testing group. You post things within an hour of them going on the official forum and you post here continuously throughout the day.
Why are you not a tester? Dont you want the game to be the best it can be?
I do test quite a bit on PTU, but i did not apply for Evocati. Not enough time for that. Plenty of time later when Alpha 3.0 is on PTU.
Reading news and posting is easy on a smartphone or tablet, almost anywhere, almost anytime. Testing is not - on a smartphone :-)
And lets not forget - Evocati are under NDA. They could tell us, but then they have to shoot us ;-)
The format of the report was changed as said 2 weeks ago to properly display all that's left before 3.0 reaches the public PTU, instead of accounting for the next phase of the Evocati (the shopping/cargo/trade pass they've been doing now mostly complete already focus on missions).
So remaining category has its set of remaining bugs and ongoing tasks to be completed:
Shopping
Missions
Ships/Vehicles
Traversal
MobiGlas
Performance and Stability:
This makes up 57 Bugs and 254 ongoing Tasks. If you are curious about the high number of tasks mind this is everything issued up to do, from adding a texture to some object to implementing a new feature; the bug fixing process is the priority but the devs that aren't involved with such process will continue working on the build.
In the report page, you'll see a status update in each of those sections.
However I feel they would be better reporting progress based on whether a bug or task is finished or not.
As it stands they are basing it on ... how many hours or maybe days to finish? For example: Traversal Bugs of which there are 2 to fix is showing as 8% complete. There are methods that can be used to underpin percentages but given their track record if they do its not working well.
They would probably be better adopting such a "0/100".
Arguments can be made that its "pessimistic" and on the plus side its very, very simple; something is done or its not. The key is to ensure scheduled tasks are not to long - aim for 4 week max say but allow up to say 8 weeks to ensure tasks are tied to actual "deliverables".
If under reporting their actual progress (albeit slightly) meant that the target dates they gave out were a little pessimistic well I don't think that would be a bad thing.
Edit: especially as their resource management hasn't gotten any better.
Traversal <snip>
Gravlev Polish and Tuning hasn’t been started yet, as the designers and programmers that will be looking at this have been working on other 3.0 features that are at a less complete state.
So they didn't know the designers and programmers scheduled to do this had that to do? Sigh.
Most of the missions for 3.0 are fully functional and any that require bugfixing and polish will be done in the remaining weeks. A few more missions are being worked on until mid-November, so there is enough time for thorough testing.
Worked on until Mid - November. This suggests to me a target - ! - of getting to PTU for Thanksgiving (i.e. Thursday 23rd November for people outside the US). I know people have "doubted" that it will be many weeks or months before it gets to PTU but for me this ignores a) the pressure and b) the fact that PTU is just another test server. So things don't need to be finished finished just good enough - same as was the case with Evocati release.
And that would maybe suggest a Christmas release to general.
As it stands they are basing it on ... how many hours or maybe days to finish? For example: Traversal Bugs of which there are 2 to fix is showing as 8% complete. There are methods that can be used to underpin percentages but given their track record if they do its not working well.
I think you misunderstood. That 8% is not completion, it's percentage of total issues in that category.
Traversal has 24 total issues, 2 of them are bugs, so 2/24 = 8% bugs. The rest in traversal are tasks, so 22/24 = 92% tasks.
So they didn't know the designers and programmers scheduled to do this had that to do? Sigh.
No, as the schedule flows if the devs that had to tackle that feature fall behind schedule to work on other stuff then forcefully that work doesn't start until they are free to work on it, causing the push back.
It's about the availability of resources, and with the whole bugfixing process not any surprise devs need to go back to something they already finished to support bugfixing and other stuff.
So they didn't know the designers and programmers scheduled to do this had that to do? Sigh.
No, as the schedule flows if the devs that had to tackle that feature fall behind schedule to work on other stuff then forcefully that work doesn't start until they are free to work on it, causing the push back.
It's about the availability of resources, and with the whole bugfixing process not any surprise devs need to go back to something they already finished to support bugfixing and other stuff.
No Max. This is a weekly update. Something CR (almost head in hands) has alluded to with comments in the run up to the Evocati release along the lines of "if it can't be finished on time why don't they just say".
I know how it works. I have managed projects from small to larger than SC. Software and non-software. In multiple EU countries, in the US, cross-continent.
At the end of the day the "not knowing" comes down to those involved believing that pigs will fly rather than saying this will probably need more time. And if it doesn't then it finishes a few days early! As I said its a weekly update. It can be a very tough thing to instill though because people - by default - don't like reporting slippage until they have to.
And for management as early as possible is best. CR would be much not having to explain slips and would much rather report stuff happened on time or even early!
Comments
The Alpha 3.0 Production Schedule got updated:
Burn Down, current status (Weekly Update):
The number of issues lies in 23, mind these are the list of bugs to deal with before pass update on the Evocati build that should focus on fixing/polishing the Shopping & Cargo Features.
Completed this past week:
Resuming last week:
- 1 Completed Tasks
- A total of 5 features left.
Bug-fixing changelog for the past week in the full schedule page and in Evocati patch notes of the pretty much daily builds.
The Alpha 3.0 Production Schedule got updated:
Burn Down, current status (Weekly Update):
What's still in progress:
Bug-fixing changelog for the past week in the full schedule page and in Evocati patch notes of the pretty much daily builds.
They keep the numbers the same, but when completing some feature takes too long, they just move some of the planned features to a higher version number.
If you do this gradually enough , like over a year, little by little moving over features to future versions, it almost seems as if they stay close on schedule In version numbers anyway, not in actual planned features. 'Yeah, I am sorry, we have to move this feature and this feature over to next year, but don't worry, 3.1 is on schedule!'
Anyway, you could almost compare it to this company that keeps saying that there is enough money to finish SQ42, and with the box sales of SQ42 they would have enough to finish development of SC. Meanwhile they keep introducing more ridiculously priced ships that you can 'fund' the game with. Oh wait , that is the same company, doh.
I wonder if they plan to continue selling these packages after game release too. I would be in awe if they manage to do that. It would be like GTA Online, but then greatly improved. Shark cards, but then with a better pitch.
The release and completion of the game in the case of SC are not tied together either, so to be seen when they decide to call that.
The Alpha 3.0 Production Schedule got updated:
Burn Down, current status (Weekly Update):
So the schedule is going to change format next week, I guess mostly because the Evocati is working with multiple update passes to fix and polish up part by part of the build.
(currently shopping, cargo and trading, next the new mission system)
I wasnt going to bump this because predicting what these guys are going to do is about as hard as saying the sun will come up tomorrow (and it still comes up even if its cloudy).
Also the first line is pretty awesome since apparently they let peopel play 3.0 on THEIR servers and of course it ran flawlessly, yet the leaks from the avocados say 8-15 frame rates as long as you have 32 or more gig of Ram and the best vid card money can buy.
That is what the leaks say:
"Q: How's performance in the most recent patches?
A: E and F were dogshit with constant 30K errors and CTDs. Patch G gave me 60FPS on the US server for the first time, and it didn't drop me to less than 30 FPS on a full server. "Have fun
But always nice to see what parts of guys breaking an NDA people will cherry pick and when they will cherry pick them. When the leaks were negative they were 'fake' when they put a nice spin on it theyre being quoted as 'proof'.
I can neither confirm nor deny that those claims are correct or fake.
It is your decision what you believe ... or not believe
Have fun
That's nothing but the point of ETF itself, there's a very flippant nature of this factor as testing endures, hoping they work out the stability of the changes they pushed before to maintain the network performance as they continue work.
Bottom line is they have these movies they call gameplay then they have (alleged) people 'playing' the latest build, all running nearly 100% flawlessly. When not even the staunchest white knight can claim what they offer on the test bed runs even adequately and thats the stuff thats been there for months.
You cant watch a guy that is streaming (one of the half dozen) where you will find a stream that lasts more than 30-45 minutes before a CTD or a freeze up. Unless of course its the guy standing in a hanger facing a blank wall and him pontificating about the 'future' of SC.
The 'regular' guys know where ot go and what to avoid so their crashes have been minimized but even they fall victim to them. But get a random guy trying it out for the first time or the first time in awhile and it will be constant meltdown and freezes.
Thats why I always look forward to the dog and pony show when they do these demonstrations at the conventions because its so far from reality it could probably be considered 'fake' on that aspect alone. But since it goes way beyond that in terms of 'production' value its obviously faked in more ways than that. Or the way it runs is proof that its fake. People can draw their own conclusions.
Why are you not a tester? Dont you want the game to be the best it can be?
Reading news and posting is easy on a smartphone or tablet, almost anywhere, almost anytime. Testing is not - on a smartphone :-)
And lets not forget - Evocati are under NDA. They could tell us, but then they have to shoot us ;-)
Have fun
The engine is quite robust and optimized graphically, it's the networking that hinders performance. Anyone that plays Star Citizen knows that.
You don't choose to be an Evocati, the Evocati chooses you
The format of the report was changed as said 2 weeks ago to properly display all that's left before 3.0 reaches the public PTU, instead of accounting for the next phase of the Evocati (the shopping/cargo/trade pass they've been doing now mostly complete already focus on missions).
So remaining category has its set of remaining bugs and ongoing tasks to be completed:
This makes up 57 Bugs and 254 ongoing Tasks. If you are curious about the high number of tasks mind this is everything issued up to do, from adding a texture to some object to implementing a new feature; the bug fixing process is the priority but the devs that aren't involved with such process will continue working on the build.
In the report page, you'll see a status update in each of those sections.
However I feel they would be better reporting progress based on whether a bug or task is finished or not.
As it stands they are basing it on ... how many hours or maybe days to finish? For example: Traversal Bugs of which there are 2 to fix is showing as 8% complete. There are methods that can be used to underpin percentages but given their track record if they do its not working well.
They would probably be better adopting such a "0/100".
Arguments can be made that its "pessimistic" and on the plus side its very, very simple; something is done or its not. The key is to ensure scheduled tasks are not to long - aim for 4 week max say but allow up to say 8 weeks to ensure tasks are tied to actual "deliverables".
If under reporting their actual progress (albeit slightly) meant that the target dates they gave out were a little pessimistic well I don't think that would be a bad thing.
Edit: especially as their resource management hasn't gotten any better.
Traversal <snip>
- Gravlev Polish and Tuning hasn’t been started yet, as the designers and programmers that will be looking at this have been working on other 3.0 features that are at a less complete state.
So they didn't know the designers and programmers scheduled to do this had that to do? Sigh.Missions
Worked on until Mid - November. This suggests to me a target - ! - of getting to PTU for Thanksgiving (i.e. Thursday 23rd November for people outside the US). I know people have "doubted" that it will be many weeks or months before it gets to PTU but for me this ignores a) the pressure and b) the fact that PTU is just another test server. So things don't need to be finished finished just good enough - same as was the case with Evocati release.
And that would maybe suggest a Christmas release to general.
Traversal has 24 total issues, 2 of them are bugs, so 2/24 = 8% bugs.
The rest in traversal are tasks, so 22/24 = 92% tasks.
It's about the availability of resources, and with the whole bugfixing process not any surprise devs need to go back to something they already finished to support bugfixing and other stuff.
I know how it works. I have managed projects from small to larger than SC. Software and non-software. In multiple EU countries, in the US, cross-continent.
At the end of the day the "not knowing" comes down to those involved believing that pigs will fly rather than saying this will probably need more time. And if it doesn't then it finishes a few days early! As I said its a weekly update. It can be a very tough thing to instill though because people - by default - don't like reporting slippage until they have to.
And for management as early as possible is best. CR would be much not having to explain slips and would much rather report stuff happened on time or even early!