directx

Early DirectX 12 Benchmarks

AnandTech have a preview of Microsoft’s upcoming DirectX 12. It’s a very promising start.

DirectX 12 will be the biggest iteration of the DirectX API since its inception during the Windows 95 days. It helped establish a set of standards that helped ease the pain of game development on Windows platforms. But as time marched forward and new ideas were tacked on, DirectX became bloated and we needed a new approach. AMD introduced Mantle which promised improved performance for its GPUs by granting lower level access to the hardware and minimizing the performance impact of the API.

That’s all well and good but it’s an AMD thing; what we really need is for a “platform” holder like Microsoft to overhaul things and it looks like they’ve done that with DirectX 12. And with Windows 10 being a free upgrade for the first year, nearly everyone can take advantage of the new API.

I predict the earlty performance gains will be none too dissimilar to the AnandTech conclusion to Mantle’s Civilization: Beyond Earth’s usage. Just replace Mantle with DirectX 12:

As for Mantle, it’s an interesting option if you have an AMD card. While we’re not seeing any huge benefits on our test system, it does typically run 5-10% faster than the DX11 path; this is good but in most cases it’s not enough to really make a palpable difference to the end user. The word is that lower spec CPUs like the Intel Core i3 and Pentium Anniversary Edition along with AMD APUs can benefit even more. I’m not sure how many people are actually pairing up slower CPUs with high-end GPUs these days, as $200 CPUs are pretty common for gaming systems, but additional options are never a bad thing.

Mantle does have a much more tangible impact on minimum frame rates, and this is always beneficial, particularly once a game is averaging more than 60FPS. In some cases Mantle was able to improve the minimum frame rates (e.g. on R9 290X) by 40-50%, though that was only at less strenuous settings. Still, even the 5-15% increases in minimum frame rates at higher resolutions are welcome.

Hopefully we’ll see more substantial gains down the line.

Re: Wireless Bridging & Forking Custom Firmwares

I’m no stranger to forks of custom firmware. Hell, I used Shibby’s Tomato custom firmware for a while and that lineage is even deeper.

Linksys Linux based firmware -> HyperWRT -> Tomato -> TomatoUSB -> Shibby’s Tomato build

I installed Shibby’s Tomato on my ASUS RT-N66U for the longest time but decided to switch over to a spin-off of the official firmware known as ASUSWRT-Merlin. (Link req’d) With the spin-off (known as John’s fork), the ability to crank up the transmit power on the wireless radios was restored and what a difference it has made. I can nearly get away without a bridging/repeating solution.

As I alluded to yesterday, I currently have the RT-N16 acting as a wireless repeater to the RT-N66U’s 2.4GHz radio. The RT-N16 sees the 2.4GHz network at 100% signal strength but the performance fluctuates wildly at times. Some times even going down to 1 Mbps. I’m not sure what’s the culprit here. Is it the RT-N16’s hardware or firmware? I know the RT-N16’s 2.4GHz radio isn’t as powerful as the RT-N66U’s but at 78-83% signal strength is still pretty good.

On top of performance issues, the Roku 3 cannot connect to or acquire an IP address through the wireless repeating solution. I believe it has something to do with the fact that the RT-N16 isn’t acting as proper bridge. Every other device functions fine but the Roku 3 is finicky like that. Thankfully, it’s close enough to the RT-N66U to use the 2.4GHz radio.

So what are my options at the moment?

Let’s get the obvious out of the way: Hard wiring — Ethernet, power line or otherwise — isn’t possible which is why I’m investing so much time and effort into this solution.

1. Swap out the RT-N16 and try the Linksys E4200 w/ Shibby’s Tomato installed and try to establish a Wireless Ethernet Bridge
2. Change the RT-N16’s firmware back to the latest Shibby Tomato build and try to establish a Wireless Ethernet Bridge
3. Buy an ASUS RT-AC68U ($230) and couple it with my RT-N66U and form a Media Bridge (with ASUS’ firmware) or Wireless Ethernet Bridge (with Shibby’s Tomato). I can also try wireless repeating again.
4. Buy a Netgear R7000 Nighthawk ($180) and couple it with my RT-N66U to form a Wireless Ethernet Bridge or try wireless repeating again.
5. Buy an Amped SR20000G ($130) or Netgear EX6200 ($130) — actual wireless repeaters.

Plenty of options and I still haven’t exhausted the free ones. This has been somewhat enjoyable but at the same time, I wish everything would work as I wanted.

Random Thought of the Week

Pull ups are a real challenge. I will persevere though.