you might have some advice to give on how to be insensitive

Caught via mention on Nicholaz Beresford's blog:

Dave L. said:

A number of people have been e-mailing me directly complaining that this doesn't work with their GeForce 7 GPU. GeForce 7 was introduced in June 2005, which means it would be starting preschool in the fall. It's old enough to talk in complete sentences, it's beyond obsolete. Frankly, drawing the line at GeForce 8 (not even current gen any more) for a still-in-development high end feature is generous. That would be like Crytek saying they're not going to add any features to Crysis that are DirectX 10.0 only (probably a bad example what with the whole Vista thing, but you see my point). I'm not going to apologize for it, and I'm not going to pander to people who think a 3 year old architecture should receive as much support and attention as current offerings.

The way I see it, if you don't care enough about graphics to keep and maintain an up to date system, you shouldn't expect to experience significant graphics improvements. If I invest resources to supporting an old chip that's being phased out, I get a much lower return on that investment than if I would have put those resources into a chip that's becoming mainstream. The number of installed GeForce 7's in the world is decreasing while the number of GeForce 8's is increasing, and I don't expect GeForce 7's to make a comeback.

Now, before the over reaction machine spins up and I get an e-mail about an open letter from the cheapskates of the world, this doesn't mean Second Life is going to be GeForce 8 only, just that you'll need that kind of card if you want the absolute best experience. Second Life will still run on that POS laptop your boss let you take out of the recycling bin at work.

This strikes me as insanely dismissive, and I wasn't the only one who saw it that way. Callum L. responded:

I think some thought needs to be put in to what exactly Second Life's target market is. I don't know what goes on inside the LL wall but I've not seen any such discussion outside it.

I see it as a "casual game". Its not for the hardcore gamers with seemingly unlimited disposable income. They're all playing WoW, Counterstrike, Call of Duty, Halo and whatnot. Second Life is the virtual world for the rest of us. [It's] for the casual computer user, who buys a new mid-range system every 5 years. [It's] for that family member who's using that spare box you gave them.

If you accept this view, this means LL must be much more conservative about pushing minimum requirements forward than most games. The target should be a mid-rangeish system from 5 years ago, the client should still run nicely on such a machine. All LL developers should be forced to do all testing on such a machine unless they're working on the new shinies.

Which isn't to say LL shouldn't be trying to keep the graphics engine current. It just means effort has to be put in to "graceful degredation". Or perhaps "progressive enhancement". All it takes is to bother to put the effort into it...

And on the heels of that response came the one that changed the name of the thread. Matthew D. said:

I'm changing the title of this thread since I have no problem with DaveL restricting himself to high-end graphics hardware for what is, let's face it, an early preview of an experimental feature in an experimental source code branch which isn't even scheduled to go into the release branch yet (and may turn out never to be released!)

I do have issue with the "if you[r] machine is older than a few years, stop being a skinflint and buy a new one" implication of the e-mail though!

There is a perception that LL is concentrating on fancy features which require top end, up to date machines at the expense of working on improving the stability and performance of SL on mid-range/older machines. DaveL's comments, and observations such as below about SL's performance on lower end machines (and there seems to be some controversy as to whether WL with all the shaders turned off runs better or worse than pre-WL on older hardware), only re-inforces this perception.

To be honest, there is no problem with LL concentrating effort on high-end machines *if* their target audience is just the high end gaming community. However, if the target audience are typical home users (who are likely to buy mid-range machines which may even today come with the sort of hardware DaveL would regard as dead; and unlikely to be replacing them as frequently as 3-4 years), educational users (again buying mid-range machines, and typically laptops with embedded graphics), business users (typically embedded graphics and typically laptops) then needs to put most of its effort into getting it running well on the hardware such audiences will typically have - not berating them for not having the latest and greatest (but this shouldn't stop a few programmers doing high end stuff in experimental branches....).

This is very much the essence of the Gartner report earlier this year - the typical business user does not run SL well - and for LL to say "tough - upgrade all your machines", is a little like the tail wagging the dog if LL wants to make inroads into that market, and they'll supply respond "fine - we'll use something else". That would be fine if LL has no interest in that market - however, the whole secondlifegrid thing suggest that they *do* wish to attract those users!

And they do. The whole upcoming "Your World, Our Trade Demo" (such a lovely catchphrase!) seems to be geared towards emphasizing business concerns over the casual populace. Which, I suppose, is fine--if that's the audience they're after.

And as Matthew points out--if that is their target, now? They need to realize that, and react in kind. By not stripping away lower-end features that will gut that sector of the population, time and time again. Forget the casual computer user--the average businessman in the cubicle won't come in. And all the corporate kowtowing in the world, at that point, won't make them--if they can't play on the machines they have.

Nicholas C. responds:

Matthew Dowd wrote:
^ To be honest, there is no problem with LL concentrating effort on
^ high-end machines *if* their target audience is just the high end gaming
^ community. However, if the target audience are typical home users (who
^ are likely to buy mid-range machines which may even today come with the
^ sort of hardware DaveL would regard as dead; and unlikely to be
^ replacing them as frequently as 3-4 years), educational users (again
^ buying mid-range machines, and typically laptops with embedded
^ graphics), business users (typically embedded graphics and typically
^ laptops) then needs to put most of its effort into getting it running
^ well on the hardware such audiences will typically have - not berating
^ them for not having the latest and greatest (but this shouldn't stop a
^ few programmers doing high end stuff in experimental branches....).

Here here. One thing I keep hearing about There is how it can run on a 56K modem. Not that I think we want to go THAT far, but if we ever want SL to be ubiquitous, it should be usable on a machine that was made at least 3-4 years ago. And even then you're letting quite a few people out. I have several friends who I keep trying to introduce to SL, but we just can't get it to run on their machine.

Heck, I myself signed up for an account back in 2004 but the software wouldn't run on my machine. It was 2 years before I came back. At that point 2 years didn't mean much, because there weren't too many competitors. Now 2 years is the difference between success and marginalization. (Is that a word?) The point is that we're getting people use to a 3D world; if they can't run SL, they'll find something else, and they won't be back.

---- Chase

Nicholaz, in fact, states in the post addressing this issue:

Now today comes an interesting comment from the Lab which is interesting. What it essentially says is, that except for the lowest end, in WL there are no settings which exactly match the older viewers. Of course you can turn off the sky, but a bunch of new features come in a packet (Preferences, Graphics, Custom, Basic Shaders), which will either hit your performance or which will leave you with noticeably less appealing visuals. I.e. with mid-range machines, you can trade off graphics or trade off performance, but there is no way in the middle as with old viewers. (If you are getting funny results with Eye Candy, even while you think you have similar settings, the Basic Shaders option may be the key).

This may be a pesky little technical detail, but I found myself being irritated by the Linden's claim that Windlight was optional and could easily be turned off to give you an experience as before and the fact that whatever I did, I either lost visuals or framerates.

So now, with the new mandatory download candidate, there's no way to gain effective performance on slower machines, is what it sounds like. Yet another case of Linden Labs being woefully out of touch with their base population.

I became a paid user for this? *stares dolefully at her 2005-era machine*

Not only that, but now? Every time I log in--using my Nicholaz browser--the login screen screams at me to update to the latest candidate, but won't give me an active link on how to do that!

Gah.

And more activity on the JIRA issue that supposedly closed--this time backed up by Shoshana Epsilon, the head of the SL5B celebration's art department: now, apparently, no child avatar can be photographed with any adult avatar.

Not only is this deeply stupid, but it's also from the same Linden on crack, Everett. Someone really needs to drop-kick his av down a well for the next month or so, give him some quiet alone time to reflect on how much of a bonehead he's been through this entire thing.

Let's put it this way--though I've never gone out of my way to mention it here, Frontier Linden has always had the rating of 'most useless human ever' on my personal list, based on an incident in Rivula, where he came in and--after being told exactly what the sim managers had done to get to the point of calling a Linden based on tanking sim performance--insisted that the reason the sim was failing, was that the sim manager hadn't cleared his cache enough. Like, suggested cache-clearing as the answer for sim issues nine times in a half-hour conversation.

"Have you cleared your cache?"
"Yes."
"Well...you know, if you cleared your cache--"
"We've done that."
"Hmm...well, you know, many people forget to clear their cache--"
"CLEARED. THE CACHE. What else?"
*sound of Frontier pausing to think*
"Have you tried clearing cache and relogging?"

Now, he redeemed himself partially during the doll protests--mainly, for being willing to talk to the group of us when Torley was, hmm, shall we say, 'unavailable' during 'available' office hours for a scheduled meeting--and for that, I backed off a bit in personal conversations.

Frontier? At this point looks like a brightly shining beacon of Linden competence compared to Everett.

I'm apparently going to be aiding one of the builders on the SL5B project--text-based capacity, only, not actual *building* of anything--so I'm not on the SL5B group. They are, however, and it's interesting hearing what's coming from the project organizers, day by day...

I'm fairly sure SL5B has officially reached "shambles", but has not quite reached "catastrophe".

Wait a week; we could be moving (at speed) in that direction.

(Vint Falken's comment on the new rumors? Succinct and to the point as usual.)