SG14 Games Dev/Low Latency Plans and logistics
已查看 114 次
跳至第一个未读帖子
michaelw
2015年8月4日 15:25:472015/8/4
收件人 SG14 - Game Dev and Low Latency、unofficial-r...@googlegroups.com
Hi all, my name is Michael Wong. I am the Chair of SG14 and the Canadian C++ Standard rep. First, I like to ask anyone still on unofficial cxx real time google group to transfer to SG14, to continue the discussion and development of SG14 Games Dev/Low Latency.
Since I asked about who are represented in this post:
https://groups.google.com/forum/#!topic/unofficial-real-time-cxx/wdCn14UTISQ
You can probably find a bit about me through my isocpp.org profile or my blog here which talks about the background and history of the creation of SG14:
http://wongmichael.com/2015/07/22/on-gamessg14-and-tmsg5-from-the-view-at-the-may-2015-c-standard-meeting-in-lenexa/
I worked with Nicholas Guillemot and Sean Middleditch over the last couple weeks to plan how to organize this group.
Our plan and usual work mode in C++ Standard is to have 1- 6 like-mind and common interest people get together to work on specific proposals by first getting together talking (by telecon, email, or meeting), write a paper down their ideas, present it to the SG at their telecons/meetings, iterate until all objections are removed (possibly passing it through an umbrella committee as well) , then proceed to converting them into C++ Std words (called Standardrese), then decide how to publish it (through a Technical report or directly into an International Standard).
With that said, I like offer some thoughts on how we can achieve that process.
1. Official Face-2-Face meetings
Yes we have at least 2 of them. In fact, the first one is coming up very fast, on Wednesday of CPPCon on Sept 16.
A second one will occur during week of GDC 2016 March 14-18th in SF at the Moscone Center with the sponsorship of Sony (they have committed to pay for the room and breaks, I believe). If you need justification to attend GDC's SG14, consider giving a talk as the Call for submission just opened:
The deadline is somewhat soon: Thursday, August 27
For more: http://www.gdconf.com/conference/c4p/
Otherwise, if you need to convince your boss and think an official invite from me will help, I can also provide that on demand.
First, about CPPCON, we will have a room at The Meydenbauer Center to have a full day meeting to review the papers. We will have a signup (working on that right now) as we only have space for about 50 people. This is a working meeting so only people who really want to participate in reviewing and offering feedbacks to papres, or have proposals should attend. We can't stop people from stopping in and out, but paper authors will need to know to be present to give their proposal.
CPPCon and ISOCPP.org foundation has already agreed to pay for the room, and break which will just piggy-back on CPPCon. We are working on what it means to have people who just want to attend SG14 and not CPPCon. On Wednesday, there will be no Games Track talks.
What are Games Track Talks? we have identified the following as likely Games Track talks. I have to contact the authors to have them agree to have them tagged as such. But one effect of that is that Games Track talks will not occur on Wednesday to give the authors time to attend SG14.
Currently these are the Games Track Talks on CPPCON. If you see others that I have missed or disagree with these choices, please reply and I can check it out.
In order to have a useful meeting, we need people to submit their ideas, proposals well before the 2 SG14 meetings. That mean we will have SG14 Paper submission deadlines (just like a normal C++ Std meeting). These deadlines enable papers to be reviewed before the meeting by others and meaningful feedback be given at the meeting.
I don't know if this kind of rigorous procedure will work, but I like to propose we try it out first, because these are very similar to how C++ Standard meetings work. If it doesn't then we can resort to another form which I have run in SG5.
Given CPPCon's close proximity, the paper mailing deadline is also very soon. It will be the end of the month on
Friday, Aug 2 giving us 2 weeks until CPPCon.
A similar paper mailing deadline will also exist for GDC 2016, being Feb 26th, 2016.
I will discuss the mechanism and the expected content of the papers in another post on in a follow-up telecon. But the general idea is to commit your design (after consultations) down to paper. You may summarize the reflector discussions, and offer your opinion on which design to go with, you may also add code examples with the aim of getting feedback from the group. On the other hand, some proposals can be just a barebone wishlist, or meta level idea. Others, especially very tiny fixes can be direct changes to the wording of C++ Standard.
Note that these papers will ultimately be submitted to the WG21 C++ reflector mailing lists, can can be VERY helpful in forming future talks for CPPCon and GDC.
3. Should there be SG14 working telecons in addition to theF2F meetings? I run them every other week in SG5 Transactional Memory, but at this point, I am not sure we are ready for such rapid pace. I like to propose once a month, with the purpose of using it to discuss some logistics, but mostly be a focal point for technical discussions with the aim of pulling in people of like-mind to work on proposals.
I like to propose August 12 as an Offiical SG14 telecon to start and see how it works out. I realize not everyone will be able to make it, and given the world-wide interest, I usually aim the call at 12-1 ET (9-10 PT, 5-6 UK, 6-7 Europe) which gives often the broadest time zone coverage.
What to talk about? for this first call, I would set some logistics discussion time as not everyone is familiar with C++ Std workings. But I also want to give some time to organize groups to work on immediate paper topics such as what has been discussed so far
If more telecons are needed before the Aug 28 deadline, then we can organize another one.
4. Scope of SG14
I like to have a discussion about the scope of SG14 in a different post. Let me do that in the next post.
Please let me know what you think about these logistics.
Since I asked about who are represented in this post:
https://groups.google.com/forum/#!topic/unofficial-real-time-cxx/wdCn14UTISQ
You can probably find a bit about me through my isocpp.org profile or my blog here which talks about the background and history of the creation of SG14:
http://wongmichael.com/2015/07/22/on-gamessg14-and-tmsg5-from-the-view-at-the-may-2015-c-standard-meeting-in-lenexa/
I worked with Nicholas Guillemot and Sean Middleditch over the last couple weeks to plan how to organize this group.
Our plan and usual work mode in C++ Standard is to have 1- 6 like-mind and common interest people get together to work on specific proposals by first getting together talking (by telecon, email, or meeting), write a paper down their ideas, present it to the SG at their telecons/meetings, iterate until all objections are removed (possibly passing it through an umbrella committee as well) , then proceed to converting them into C++ Std words (called Standardrese), then decide how to publish it (through a Technical report or directly into an International Standard).
With that said, I like offer some thoughts on how we can achieve that process.
1. Official Face-2-Face meetings
Yes we have at least 2 of them. In fact, the first one is coming up very fast, on Wednesday of CPPCon on Sept 16.
A second one will occur during week of GDC 2016 March 14-18th in SF at the Moscone Center with the sponsorship of Sony (they have committed to pay for the room and breaks, I believe). If you need justification to attend GDC's SG14, consider giving a talk as the Call for submission just opened:
The deadline is somewhat soon: Thursday, August 27
For more: http://www.gdconf.com/conference/c4p/
Otherwise, if you need to convince your boss and think an official invite from me will help, I can also provide that on demand.
First, about CPPCON, we will have a room at The Meydenbauer Center to have a full day meeting to review the papers. We will have a signup (working on that right now) as we only have space for about 50 people. This is a working meeting so only people who really want to participate in reviewing and offering feedbacks to papres, or have proposals should attend. We can't stop people from stopping in and out, but paper authors will need to know to be present to give their proposal.
CPPCon and ISOCPP.org foundation has already agreed to pay for the room, and break which will just piggy-back on CPPCon. We are working on what it means to have people who just want to attend SG14 and not CPPCon. On Wednesday, there will be no Games Track talks.
What are Games Track Talks? we have identified the following as likely Games Track talks. I have to contact the authors to have them agree to have them tagged as such. But one effect of that is that Games Track talks will not occur on Wednesday to give the authors time to attend SG14.
Currently these are the Games Track Talks on CPPCON. If you see others that I have missed or disagree with these choices, please reply and I can check it out.
- Definitely games related:
- C++ for cross-platform VR development
- Testing Battle.net (before deploying to millions of players)
- The current memory and C++ debugging tools used at Electronic Arts
- our own SG14 talk
- Probably interesting to games developers:
- Live lock-free or deadlock (practical Lock-free programming)
- Reflection techniques in C++
- (Maybe?) Cross-Platform Mobile App Development with Visual C++ 2015
- How to make your data structures wait-free for reads
- C++11/14/17 Atomics the Deep dive: the gory details, before the story consumes you! (by Michael)
- C++ Atomics: The Sad Story of memory_order_consume: A Happy Ending at Last?
- C++ in the Audio Industry
- 3D Face Tracking and Reconstruction using Modern C++
- Implementation of a component-based entity system in modern C++14
- Probably less interesting to games developers:
In order to have a useful meeting, we need people to submit their ideas, proposals well before the 2 SG14 meetings. That mean we will have SG14 Paper submission deadlines (just like a normal C++ Std meeting). These deadlines enable papers to be reviewed before the meeting by others and meaningful feedback be given at the meeting.
I don't know if this kind of rigorous procedure will work, but I like to propose we try it out first, because these are very similar to how C++ Standard meetings work. If it doesn't then we can resort to another form which I have run in SG5.
Given CPPCon's close proximity, the paper mailing deadline is also very soon. It will be the end of the month on
Friday, Aug 2 giving us 2 weeks until CPPCon.
A similar paper mailing deadline will also exist for GDC 2016, being Feb 26th, 2016.
I will discuss the mechanism and the expected content of the papers in another post on in a follow-up telecon. But the general idea is to commit your design (after consultations) down to paper. You may summarize the reflector discussions, and offer your opinion on which design to go with, you may also add code examples with the aim of getting feedback from the group. On the other hand, some proposals can be just a barebone wishlist, or meta level idea. Others, especially very tiny fixes can be direct changes to the wording of C++ Standard.
Note that these papers will ultimately be submitted to the WG21 C++ reflector mailing lists, can can be VERY helpful in forming future talks for CPPCon and GDC.
3. Should there be SG14 working telecons in addition to theF2F meetings? I run them every other week in SG5 Transactional Memory, but at this point, I am not sure we are ready for such rapid pace. I like to propose once a month, with the purpose of using it to discuss some logistics, but mostly be a focal point for technical discussions with the aim of pulling in people of like-mind to work on proposals.
I like to propose August 12 as an Offiical SG14 telecon to start and see how it works out. I realize not everyone will be able to make it, and given the world-wide interest, I usually aim the call at 12-1 ET (9-10 PT, 5-6 UK, 6-7 Europe) which gives often the broadest time zone coverage.
What to talk about? for this first call, I would set some logistics discussion time as not everyone is familiar with C++ Std workings. But I also want to give some time to organize groups to work on immediate paper topics such as what has been discussed so far
- flat_map
- fixed point
- uninitialized algorithms
- string stuff?
- intrusive containers?
- other things?
If more telecons are needed before the Aug 28 deadline, then we can organize another one.
4. Scope of SG14
I like to have a discussion about the scope of SG14 in a different post. Let me do that in the next post.
Please let me know what you think about these logistics.
回复全部
回复作者
转发
0 个新帖子