Squiggler

I don't have any sources for my conjecture that bandwidth limitations impose the most severe penalty during graphically intensive scenes (which unfortunately are when fps are lowest to begin with). That's just my own conjecture. As for techpowerup's and tomshardware's scaling analysis, you can find them by googling "site name<site name=""> scaling analysis". Most important are page 23 of the techpowerup scaling analysis and the sixth page at tomshardware. I'd copy and paste the links except... for some reason I can't copy and paste links in the reply box. With regard to my conjecture, the optimal data would be an fps-over-time graph, though I guess just recording the min fps might be easier.</site>

After looking at scaling analysis on tomshardware and techpowerup I'm feeling optimistic about the bandwidth limitations of TH05/a 2.0 x2 link and especially about TH05's ~ x4 2.0 successor. That said, I have a question and a concern. First, techpowerup's scaling analysis shows that even though the 7970 and gtx 680 perform similarly when given abundant bandwidth, the 7970 does far better when bandwidth is limited. Does this difference hold between ATI and NVidia in general (if you ignore Optimus compression of course)? Second, I am concerned that the bandwidth limitations of egpu solutions may have the largest negative impact on fps during the most graphically intensive moments. Since these moments have the lowest fps to begin with, the average fps comparisons that tomshardware and techpowerup have done may cover up moments of extremely low fps. Does anyone know anything about this?

I've been following this whole egpu adventure for a while and I just saw that there is now a thunderbolt solution available from bplus for (relatively) cheap. Now I just want to make sure I understand the details of the current state of egpus. This thunderbolt solution is limited to only 10 gbps and so has the same performance as 2.0 X1 optimus, correct? Furthermore, the only current way to beat the performance of these two solutions is to either buy a ridiculously expensive solution from sonnet, magma, etc., or to get a tricky x4 link going. Furthermore, bplus' thunderbolt solution has only a modest performance penalty compared to 2.0 x16. Is that all correct? Also, are there any estimates about when bplus might come out with a product that taps into more of the bandwidth of thunderbolt?