B3D status update (AKA we're not dead yet)

AlexV

Heteroscedasticitate
Moderator
Veteran
In case the title wasn't suggestive enough, let's clarify the purpose of this thread: it's a way to bring everyone up to speed with regards to what's been happening behind the scenes at Beyond3D. Whilst you're probably to say "Easy - nothing has been happening", rest assured that's not really the case, and in fact we're making strides towards breaking our current an article a quarter/year/universal cycle model, and perhaps bring back the glory days when we were as active as Maradona on the sidelines.

For the last couple of months we've been working on a DirectX 11 testing suite, which has recently gone from "here be monsters" status to "mostly functioning". We've really tried to look at more things beyond the usual fillrate and eventually ALU tests you may see - it should be pretty clear to everyone by now that in this day and age there are oodles more factors at play in a GPU. You're likely to be seeing at least some results from this one in your next article, with development continuing.

Also, the recent revival of the eternal "canned" versus "real-world" benchmark debate(let it be known, in our humble opinion, that's like debating the merits of poking your left eye out versus poking the right one - ultimately it's still a reasonably bad thing), has made us look into solutions for doing something actually interesting when it comes to benchmarking, and what we've arrived at is likely to be somewhat novel and appealing to the traditional B3D attendance.

Those two things, as well as quite a few others, have been keeping us busy lately - and we're strictly talking about B3D things, mind you. Getting to more tangible stuff, we're tentatively aiming for a July aquiring of one Fermi SKU or another, so an article covering nVidia's latest architecture is likely to happen either this month or in August (and you can hold us up to that promise!). So, in brief, we're not dead, we have new tools, and new content is in the pipe and incoming.

Forum wise, there's also a bit to talk about. The feedback thread and Site Feedback in general have been monitored and a round of fixes and improvements to the look and feel will take place in the next month or so, to give things a kick. We'll also introduce a light-on-dark alternative theme that harks back to the really early days, a lite theme for mobile devices, and we're also starting to plan an eventual upgrade to version 4 of vBulletin which we'll ask for feedback on nearer the time.

We're also conscious there's been an increase in spam recently, so we'll be battling that on two fronts with more automated detection and blocking, and also easier tools for the moderation team to use when fighting it. They're doing a sterling job right now, so we'd like to thank them publicly for the effort. As far as a reorganisation goes, including a prune or consolidation of some of the less active forums, we're looking at options and suggestions there. Stay tuned.
 
Nice to see an update from you folks here, been a little unnerving of late wondering if this fantastic place was still going somewhere, or at least staying!
 
Nice to see an update from you folks here, been a little unnerving of late wondering if this fantastic place was still going somewhere, or at least staying!
It's been hard to spend time on it since I joined ImgTec, but it's still alive and Alex and I are starting to spend the time together working on moving things forward again. The end game for me is obviously to hand over ultimate control to Alex when the time's right, and I'll step sideways and back a bit when we get there.

All of the recent work is to facilitate that and give Alex the best possible opportunity in the long run, so it's exciting to finally be getting back to work on things to do that.
 
For the last couple of months we've been working on a DirectX 11 testing suite, which has recently gone from "here be monsters" status to "mostly functioning". We've really tried to look at more things beyond the usual fillrate and eventually ALU tests you may see - it should be pretty clear to everyone by now that in this day and age there are oodles more factors at play in a GPU. You're likely to be seeing at least some results from this one in your next article, with development continuing.

Also, the recent revival of the eternal "canned" versus "real-world" benchmark debate(let it be known, in our humble opinion, that's like debating the merits of poking your left eye out versus poking the right one - ultimately it's still a reasonably bad thing), has made us look into solutions for doing something actually interesting when it comes to benchmarking, and what we've arrived at is likely to be somewhat novel and appealing to the traditional B3D attendance.

That sounds very promising. I look forward to it.
 
Are you guys going to do a GF104 vs Cypress architecture analysis? It seems like the very best way to kick off things considering you have ~same transistor count, ~same die size on the same process node. Also the GF104 is pretty cheap!
 
Yes, but we'll probably include GF100 too. We've already been testing that chip (and Cypress) with some new tooling, with good success (mmm, 2B+ tris/sec), so the plan is to get it out ASAP while people are still interested.
 
It's been hard to spend time on it since I joined ImgTec, but it's still alive and Alex and I are starting to spend the time together working on moving things forward again. The end game for me is obviously to hand over ultimate control to Alex when the time's right, and I'll step sideways and back a bit when we get there.

All of the recent work is to facilitate that and give Alex the best possible opportunity in the long run, so it's exciting to finally be getting back to work on things to do that.

Oh, i didn't know that. Stupid question, when we will see a PowerVR chip for the pc gamers :p

More seriously, good luck to all for the website, we will be there anyway :)
 
I'm hoping you gentlemen will be pleased to hear that since Friday I have my very own 470 and have started work on the article(s).
 
How's that Fermi article coming along? Did you guys get a 460 too?
 
So... whistles...
Anyway don't kill of the ethics of technology forum. I know it doesn't get enough use, but it should :)
 
Back
Top