Jump to content
  • entries
    422
  • comments
    1540
  • views
    490582

The MVP Global Summit

Sign in to follow this  
jollyjeffers

392 views

Day One : Tuesday 27th September 2005
Arrived at London Heathrow's terminal 4 check-in area at around 10.30am. Decided to get the early train over from central London so as to guarantee no rush, especially if any trains were late/cancelled.

Hung around for a while until check-in opened at 11.45am. Got quite thoroughly bored. Queued for a good 45mins before I got to the check-out desk and managed to meet a few other MVP's waiting in the same queue.

When I get to the check-in desk it seems they've "drastically" oversold the flight. Fine by me, I got upgraded to "World Traveller Plus" (a.k.a. Economy class with marginally nicer seats).

Get through check-in and security fairly quickly. Wander round to the 'spoons in the departure lounge. Spot the only group that are occupying several tables as being MVP's. Grab a drink and order some lunch and sit down for a chat. Happened to sit down pretty much opposite the legendary Richard "superpig" Fine.

Due to my upgrade I end up sitting on my own for the 9hr 30min flight. Nothing much to say about it really. Got to the hotel in time to drop bags and then head over to another hotel to meet up with the other Brits and head out for a meal.

The meal was pretty good, ended up at "Rock Bottom" in down-town Seattle. Loud bunch though - but the waiter got about a 25% tip, so I don't think he'll complain [grin]

After the meal we find that the SBS (Small Business Server) bunch have started a private party in the connecting bar. Free drinks courtesy of their credit cards!

Day Two : Wednesday 28th September 2005
Had some breakfast in the hotel restaurant where I met Dustin "circlesoft" Franklin.

After breakfast the three of us wandered around down-town Seattle for a bit. Managed to catch one of the first busses from our hotel out to the Microsoft campus in Redmond and join the queues for registration.

After we've all sorted out the registration stuff we look around the exhibition in a couple of the adjacent rooms. Nothing particularly interesting in them - a few freebies to acquire and some displays to look at. A little disappointing really - even though I probably know more about Microsoft products than many other people, I don't know much about some pretty large areas. I was hoping the exhibition would be a load of "this is our department, we make this - here, have a look..."

Wednesday evening was the EMEA dinner, held out at one of the Bellevue hotels. Took bloody ages to get there ([headshake]) - but I'll spare you any talk of busses [wink].

At the dinner I bumped into Niko "Nik02" Suni, Simon "S1CA" O'Connor, Eyal Teler and Iain Downs - all other DirectX MVP's.

Day Three : Thursday 29th September 2005
Today's events were generally split into two parts - the executive sessions and the platform meetings.

The executive sessions were where the "top dogs" from Microsoft came along and talked to all 1500 MVP's for a bit. Steve Ballmer (To whom Bill Gates technically reports to) was pretty entertaining. I was most impressed with the honesty and to-the-point Q&A - I've seen enough of these sorts of things either from over-the-net stuff or during my time at IBM. Usually some random low-end staffer is brave enough to ask a vague question which is followed by a suitably generic and vague answer by an executive.

I can't remember all of the examples, but one part of his speech was about getting WinVista across all forms of devices - desktop, server, laptop, tablet, mobile etc... yet during the speech and Q&A Ballmer was using a pen and paper ([oh]). One of the MVP's stands up and says "If you think tablet/portable computing is the way forward why are you still using a pen and paper?". Not hostile, but direct nonetheless. Ballmer had a fairly good answer (something about tablets being inappropriately cumbersome for speeches), but I liked the way that real questions were asked and for the most part honestly answered.

The only other keynote that I paid much attention to was Jim Allchin's. He had a few things to say about the "behind the scenes" changes of how WinVista is better (Not sure I can comment on them here?!). He then launched into a huge number of Windows Vista live demonstrations - which was really interesting given that I've not really seen Vista in the flesh yet. Upon seeing this I'm even more tempted to dual-boot (or VPC) Beta1 on this machine and have a poke around [grin]

The afternoon was a bit disappointing, for some reason DirectX is classified as a "Windows Client" product group - so we got thrown in with other "similar" technologies for platform presentations. There was a real XBox 360 at the front, which Dustin managed to turn on - prompting a rather loud intro screen to appear on the video wall [rolleyes].

After a while some of us DX MVP's left our platform group to go sit outside and chat amongst ourselves. Primarily to carry on with an earlier discussion on the finer potentials of 'Geometry Shaders'...

Instead we gatecrashed the "Windows SDK" platform group meeting, which we thought was fair enough given that we deal primarily with the DirectX SDK. Turned out to be a really great meeting to attend - there wasn't really a presentation, just a debate/discussion between MVP's and the relevant SDK people from MS.

I guess I'll have to leave it as "watch this space" w.r.t. what was actually discussed. Some of the ideas and options being discussed were blatantly pipe-dreams and probably won't happen, but it's very encouraging to see that they're taking an open view as to where the SDK's (particularly versioning and distribution) is going.

Thursday evening was the "Product Group" dinner. Got some more freebies (inc. a potentially useful p2p USB networking kit). We managed to get one of the larger tables with all of the DirectX product group and saw the entirely GPU based D3D10 particle system demo (mentioned in the PDC05 slides) running on Richard Thomson's laptop. Sounds odd, but I think it was a sort of "important" moment as it was the first time many of us had actually seen the D3D10 stuff "for real" [smile]. Later on we were joined by David Weller (Community manager for WGGT, owner of the DX Developer Center).

Day Four : Friday 30th September 2005
Now this was the day that I was most looking forward to - DirectX Technical Deep Dive Sessions [grin]

We all headed over to one of the meeting rooms in the DirectX building on the MS campus and were greated by a few of the key people that actually work on DirectX.

Over the next day-and-a-half various members of the team appeared and disappeared (as well as dropped by to actually give presentations), so I can now put faces to a lot of the names that you occasionally spot on the forums/newsgroups/lists.

As much as I would like to, I probably can't go into a much detail on what was covered at the various sessions. Trying to tell which bits were public and which weren't is too tricky - and I don't wanna get busted into next week for revealing something I shouldn't have [oh]

Needless to say, a lot of it was regarding DirectX10 and it's position in/around Windows Vista. As MVP's we already have access to a lot of resources, but it was really good to be able to talk directly with the "powers that be" - ask various clarifications on points we didn't get, and comment on a few features.

Friday evening was the MVP party - hosted at the 'Experience Music Project' and Sci-Fi museum in down-town Seattle. They had a free bar. I like free bars. I leave it at that.

Sean O'Driscoll (the Executive in charge of the MVP program) hijacked a bus post-party and got it to drop off a load of us in another part of Seattle to keep the party going. I think the group as a whole proved to be good business for various bars [lol]. Some random almost managed to run off with my passport though. Bastard.

Day Five : Saturday 1st October 2005
I hope that the neighbouring hotel rooms were either empty or protected by sound proof/absorbing walls. If not they would have experienced a string of rather loud expletives due to my waking up a whole FOUR HOURS later than I'd set my alarm.

Seems that my phone (Symbian based smartphone), which I was using as an alarm clock, had crashed whilst trying to play the alarm (a slightly confused screen showing the alarm display was visible). Consequently I got a nice lie in [bawling]

So after grabbing all my stuff, still cursing quite loudly at anything/everything, I belted it down to the hotel lobby and asked them to sort out the quickest way from the hotel to the Microsoft Campus. 3 minutes later an "Executive Sedan Limousine" pulls up outside. Nice. $50 and 30 minutes later I arrive at the DirectX building on the Microsoft campus. Can't get in, due to being a saturday and no receptionist. Luckily the phone is behaving again and I get through to someone on the other side and they let me in.

I get a deserved mocking from the assembled DX staff and other MVP's for arriving a little late. Luckily I only missed the first session, and managed to arrive just in time for the start of the second.

The second session was the one I REALLY did not want to miss, so a big sigh of relief there.

"SDK Community Directions + SDK Roadmap/Samples" was an awesome 90mins of discussion/debate and general chatter. Again, I probably can't state the various conclusions or possibilities for the future... but needless to say there was definitely some potential for interesting things in the not too distant future. The ones that were of particular interest to me effectively give us DX MVP's better/more access to a number of DX resources - meaning that we might well be in a position to produce a whole lot more useful content [grin]

That led to Saturday lunchtime, which was the closing BBQ. A bit more discussion between most of the DX MVP's and a couple of the DX team - mostly as an "overflow" from the pre-lunch session.

That was officially the end of the MVP Global Summit 2005 [sad].

Friday evening ended up with me, Simon, Niko, Neil and Cheryl (2 other MVP's) sitting in the bar underneath our hotel. I spent the evening drinking and talking far too much ([embarrass]) and generally trying to get a grip on life/universe/everything in the USA. I even managed to pick up an American accent - I was vaguely aware of it, but I hoped no one else noticed. Guess I was wrong [rolleyes].

Day Six : Sunday 2nd October 2005
Despite the significant consumption of beer the night before (NOTE: American beer was much better than I was expecting) I wasn't feeling hungover in the morning. However due to the excesses of the last 3 days my IQ was just about on the zero-line, if not into the negative.

Wandered around Seattle for a bit - including a visit to the Space Needle (built for the 1962 world fair). Got back to the Hotel for 3pm and headed off to catch our flight back to the UK.

Day Seven : Monday 3rd October 2005
Just to conclude the story... ended up back at my (still packed up) flat in Nottingham around 4pm - having been travelling for roughly 17hrs. Pretty damn tired at this point, although I'm tempted to think this "Jet Lag" thing is just a private joke held by seasoned travellers.

Conclusions
  1. This is probably my longest journal entry yet. If you read this far, leave me a comment just to prove it [wink]
  2. I really hope to be around in 18 months or so for the next summit
  3. It was really good - as a social event to meet people, as a form of education and as an experience of the USA.
  4. I do think that it would probably have been much easier on myself, and more productive, if I'd had the willpower to resist the free bars ([looksaround]). Remember kids - drinking is baaaad.
  5. Deliberately pack a larger/emptier bag on the outward flight due to the inevitable hoard of corporate freebies [grin]


Any questions?
Sign in to follow this  


13 Comments


Recommended Comments

Sounds like you had fun. [grin]

Quote:
leave me a comment just to prove it


[cool]

Share this comment


Link to comment
What is an MVP? Most Valuable Programmer? (Yeah, I read all the way through.)

Share this comment


Link to comment
http://mvp.support.microsoft.com/mvpfaqs

How do you become an MVP?

Simple: help others out in online forums, newsgroups, user groups, etc, and maybe write some articles and tutorials, all without expecting any reward other than the personal satisfaction of helping others.

Seriously, none of the MVPs "tried" to become MVPs - it just happens; a number of MS staff regularly visit (and even post at) boards such as GameDev.net, they and existing MVPs do take notice of the most helpful people out there...

Share this comment


Link to comment
I'm glad you had a rather pleasant trip to the US and that you liked the beer it was Miller, right? My stepfather is a corporate guy for Miller Brewing Company, so if I don't say that he won't give me free beer for Christmas.

It sounds like an awesome experience. I'm hoping to have some similair experience in the future, maybe next years GDC.. we'll have to wait and see.

Share this comment


Link to comment
Thanks for the comments everyone [grin]

Quote:
How do I make sure I'm there next time?
Other than what Simon said, you could try gate crashing the party. But they did have a few security people around, so that might not work [smile]

Quote:
Where did Superpig touch you?


As for beer, it wasn't Miller [smile] I don't think any of them were "national" brands - all local Seattle "microbrew" - "Old Seattle", "Pyramid", "Macys" (?), "E.S.B." to name a few..

I still prefer a good English pint, but those "microbrew" beers weren't too far off.

Jack

Share this comment


Link to comment
Guest Anonymous Poster

Posted

Ap read it too!

Now are there already laptops with DX10 gpus + drivers? I can't believe it!!!

Or are they using a dx9 gpu with a dx10 driver? In which case to running completely on the GPU means its using dx9 level features only.

What dx10 feature would a particle engine need?

Share this comment


Link to comment
AP: There aren't any DX10 GPUs available yet - though it's entirely possible some IHVs already have stuff in silicon for internal testing.

With D3D you can however test new features & develop code before you have the available hardware by using the reference rasterizer, e.g. my first pixel shader 1.1 program was tested using the REF device.

As for particle systems - you don't need DX10 features for a particle system, but geometry shaders can be handy for generating quads from particles, amongst many many other cool things.


Jack: I think the jetlag is drastically reduced if you manage to get some sleep while travelling, which I noticed you did [smile] - I can't sleep on planes so for me ~28 hours on the go and jetlag left my body in a seriously disorientated state on Monday night - or maybe it's just an old age thing [wink]

Share this comment


Link to comment
Guest Anonymous Poster

Posted

Heh, okay. I got misled a bit when Jack said GPU based particle system, didn't think it could be the refrast. Yeah I know particle systems don't *need* anything - they have been done since ages. But I meant what was needed to make them really good and fast. Yeah I thought about generating the quads in the geometry shader but felt Point sprites were close to it. A completely GPU based particle system had already been shown soemtime ago (Uberflow) and collision detection on GPU has also been done. SO I was wondering what dx10 features other than render to vertex buffer, vertex texture fetch. were they using.

Share this comment


Link to comment
Quote:
What dx10 feature would a particle engine need?

There's a bit of a difference between NEED and CAN USE. You could quite easily write a particle system (I did) in D3D7-IM by rolling-your-own solution. D3D8 point sprites made it a bit easier. D3D10, via geometry shaders, allows for yet more customization. You don't need them, but if you do use them it can make your code a lot simpler and more powerful.

For any specific details you'll have to look at what the PDC slides include...

Quote:
I got misled a bit when Jack said GPU based particle system, didn't think it could be the refrast.

Whoops, my bad (Thanks for the clarification Simon!). If you look at the PDC slides (link in my earlier journal entry) they explain how the new pipeline in D3D10 allows for a completely GPU-based particle system.

What I mentioned we were looking at (on a very nice, current technology, laptop) was running on the D3D10 refrast. As is typical with REFRAST applications, it was pretty slow - but bordering on fast enough to at least see what was going on [smile]

Quote:
I think the jetlag is drastically reduced if you manage to get some sleep while travelling, which I noticed you did [smile]

Ah, okay then. I did get some sleep on the plane, but I wouldn't exactly call it even close to good sleep [lol]. I suppose I'm also a bit luckier in that I didn't have to go to work a couple of days later, and don't have any deadlines to hit at the moment...

Thanks for the comments!
Jack

Share this comment


Link to comment
You can do particle systems entirely on the GPU with PS 3.0, you just have a fixed maximum number of particles. With geometry shaders you should get a whole lot more flexibility.

Share this comment


Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!