Blur Busters Forums

lyrill wrote: ↑

23 Dec 2020, 09:01

didn't click the link but that feature was out like a year ago or something

Same difference and very likely that many other games out there still don't use this methodology as of right now.

Razer_TheFiend wrote: ↑

23 Dec 2020, 02:10

Futuretech wrote: ↑

23 Dec 2020, 00:46

Well I mean come on it's not really a "trade secret" you basically got one of the major micro-controller providers to open up the USB 3.0+ system to basically operate at a higher speed and handle it. You could have been doing this as far back as 2012-2013 when USB 3.0 popped up. Both to the psuedo- if not outright interrupt based system like found on PS/2 and also the 8,000:Hz(125 microsecond)-24,000:Hz(41.6 microsecond) properties.

I'm going to ignore the unnecessarily antagonistic tone of your message and respond to one specific point which is of interest to me - USB 3 interrupts : All literature I've read till date mentions the bus intervals for USB 3 as 125us minimum. Hell, even the "latency tolerance" is 123us.

I'm happy to be corrected and to learn, but you'll have to provide a source for your 41.67us claim.

https://www.overclock.net/threads/an-ov ... gy.1251156

wo1fwood
Registered
Joined Jan 31, 2012
83 Posts
Discussion Starter • #1 • May 1, 2012

A Special Note on Future USB Specifications
The USB 3.0 specification dramatically changes the way that a host and HID-devices communicate. The basic data flow and transfer concepts are preserved, while some packets or specifications have been integrated or removed entirely. A few notable changes in dataflow should be mentioned.

A dual simplex connection is now being used which allows packets to leave and arrive simultaneously; these transfers are also no longer serial in nature (where packets must finish before the next begins). SOP packets have been replaced by isochronous timestamp packets (ITPs), while token packets have also been removed (IN is replaced by the handshake packet, while OUT has been incorporated into the data packet). SuperSpeed USB also supports continuous bursting for all types of functional data transfers, unlike USB 2.0. Also note that in the USB 3.0 specification continuous device polling has been eliminated and has been replaced by asynchronous notices.

This last note is of particular importance as each device now sends data when it is recorded; it is still host controlled, but in a different way that no longer limits packet sizes. A SuperSpeed interrupt can move up to three maximum sized packets per service interval (125µs, formerly known as a micro-frame), meaning that the maximum theoretical rate at which data can be received by the host is every 41.6µs, the equivalent of a 24000Hz polling rate in USB 2.0 terms*. The size of a service interval (125µs) represents the minimum theoretical rate of data transmission, or what would roughly equate to a 8000Hz polling rate in USB 2.0 terms*.

* The non-serial nature of USB 3.0 makes these translations an inexact correlation, but an idea of the increase in transmission speeds can be seen through these estimations.

This report by this person came out in 2012 right around the 2011/2012 area of USB 3.0 and or more specifically the future specification process of USB 3 being released to the public.

So nearly a decade ago we could have had USB 3 devices both to mice and keyboard and preparing for Type-C in 2015. Despite Type-C not taking off and it's been 5 years and we are still dealing with A ports. Would have been nice if everything went Type-C with an adapter to A. Or at least a Type-A to Type-C adapter until Type-C took over everything.

Futuretech wrote: ↑

23 Dec 2020, 00:46

In reality your just kinda milking every minor upgrade possible. What's next 10,000, 12,000, 15,000, 20,000 and finally 24,000Hz.

In principle... I kinda agree with you about jumping the milking points to a bigger step. The 240Hz->360Hz jump should have been a 240Hz->480Hz jump, for example, to keep in line with the general "upgrade to 2x refresh rate" recommendation of Blur Busters.

But... Right now, you're looking at a pimple on a broken leg -- there are bigger nuts to crack -- a bigger more important mountain than a molehill -- and larger elephants in the same room to worry about: Systems unable to keep up with 8KHz

In other words: Some of MY Blur Busters systems is unable to keep up without re-optimizing the systems. THREE OF MY COMPUTERS, in fact, could not support 8000 Hz poll rate, until I re-optimized (e.g. installed a USB PCIe card) or re-installed Windows or other tweaks!

In some ways, this is a harder-hitting upgrade than when I first downloaded Crysis in year 2007. Bluntly... yes... a mouse that out-hammers a computer than installing Crysis in 2007? That's what I just said.

So, currently -- because I really have to optimize my systems to make the 8KHz mouse amazing -- I currently have to side with Razer here based on my mouse tests where some of my systems barely keep up. So, because of this, Razer isn't milking here; they just slammed almost a "larger-than-the-market-can-tolerate" 8x jump intead of a 1.5x jump in Hz. This is literally a "bigger-than-Crysis" upgrade to the mouse -- an 8x faster mouse.

The good news is that it's marketed to esports, people who are already on powerful GPUs and really high refresh rate monitors. You got a 1000-series GPU with an i7 and a 240Hz monitor? Good, but you only have a 50-50 chance of not having problems at 8KHz in your favorite games until you intentionally lower your Hz to 2000 or 4000.

High performance systems are somewhat more easily optimizable, but even a 3-year-old computer sometimes needs a Windows reinstall to support 8KHz -- because your system is full of computer-slowdowning crud that makes it impossible to fit 8,000 into 100% of one CPU core -- sometimes it's just easier to set the Viper to 2KHz or 4KHz and be done with it; since you get 80-90% of the 8000Hz benefits by the first 1000Hz->2000Hz jump, and you can milk most of the rest (95%) from the next 2000Hz->4000Hz jump. I can still see 4000Hz vs 8000Hz but only barely, and only on my 360Hz monitor (and not at any lower refresh rate). Though, 1000Hz-vs-8000Hz is noticeable at almost all refresh rates, if the game can keep up and the DPI is optimized.

The good news is that In theory... It's not necessary to have 24000Hz poll to get 24000Hz sensor reads.

My proposal of High Definition Mouse API (aka HD Mouse API) supports optional poll-merging, allowing 8000Hz-over-1000Hz, or 24000Hz-over-8000Hz, or anyHz-over-anyHz. This lowers USB load while improving mouse precision that allows the refresh rate race to continue towards retina refresh rates.

This is feasible because HD Mouse API mandates timestamps per position/delta report -- preferably timestamped sensor-side to overcome USB jitter which can improve/worsen when other USB devices simultaneously operate (such as a high-Hz keyboard).

Some of my computers is actually unable to keep up with 8KHz without reinstalling Windows + installing a USB PCIe card, so there are some limits-pushing. Due to large number of legacy systems, I think the Razer Viper 8KHz should default to 2000-4000Hz much like fifteen years ago where the first 1000 Hz mice of year ~2005 defaulted to 500 Hz at first.

24 KHz is really pushing it in a system's ability to keep up, and I think 100% priority should be making HD Mouse API a reality, instead of further uselessly raising poll rates! Razer liked my idea that I came up with, about HD Mouse API. I think making that a reality is WAY WAY WAY WAY more important than native 24 KHz. HD Mouse API will help solve systems unable to keep up with 8000 Hz.

For More Reading: HD Mouse API

Razer is the generous one supplying samples to seed the market. Mucho respecto. But yes, I like rising tides to lift all boats. I recommend Razer, Logitech, SteelSeries, Asus, etc, band together a consortium to push such an API forward to convince Microsoft to support it, because it will do a huge deal to mice moreso than any further upgrades beyond 8000Hz. Competitors band together all the time to create standards (example: HTML5) and this is a good opportunity.

Literally, in fact, 24KHz sensor reads transmitted at 2KHz poll (12 mouse co-ordinate sets transmitted per mouse poll), will use less system load than native 8KHz over 8KHz poll, because of the USB 8000Hz eats up 100% of a single CPU core on some systems until properly re-optimized.

Win-win-win.
8KHz with less CPU load than 2KHz!
24KHz with less CPU load than 8KHz!

Everybody (Razer and competitors) gets to sell more mice, and mainstream benefits because everything is plug-n-play while being super silky, even on potato computers. With automatic-adaptation for low performance systems, an 8KHz-capable mice would no longer be limited only to relatively clean high performance computers.

The 8KHz Viper is truly a fantastic mouse. It's 100% truly worth reoptimizing your system just to support this mouse!

It's been 15 years of being stuck at 1000 Hz. So a sudden massive 8x Hz upgrade is very welcome, even though it pushed the limits of certain computers at Blur Busters Headquarters. But this is sorely needed, and will wake up motherboard manufacturers to provide dedicated-processor USB ports specifically for the mouse, and Microsoft to optimize Windows to prevent 8KHz USB devices from pegging a CPU core to 100% and dropping polls.

Does this mouse have smoothing above 2k DPI?

Brainlet wrote: ↑

24 Dec 2020, 23:39

Does this mouse have smoothing above 2k DPI?

I am earlier told by Razer that this mouse is able to do support non-smoothed operation all the way to max DPI.

BTW, I did recommend Razer to include an "Advanced Settings" section of mouse profiles in the Razer mouse software:
- Poll Rate [500 / 1000 / 2000 / 4000 / 8000 Hz]
- Enable smoothing [On / Off]
- Enable interpolation [On / Off]
- Sensor read frequency [20KHz / 16KHz / 8KHz]

Configurable per-game.

Some games clearly work better at 2000Hz or 4000Hz, and some buggy games like Cyberpunk 2077 seems to work better at 500Hz (temporarily, until it is fixed).

Also, Razer_TheFriend, Cyberpunk 2077 isn't working well with the touchpad of my Razer Blade 15. Even though it worked well when I played a complete solo playthrough of Crysis 2 in a hammock during the summer on the Razer Blade 15 laptop. (You can't use a mouse in a backyard hammock during stay-at-home!)

Chief Blur Buster wrote: ↑

24 Dec 2020, 23:44

Brainlet wrote: ↑

24 Dec 2020, 23:39

Does this mouse have smoothing above 2k DPI?

I am earlier told by Razer that this mouse is able to do support non-smoothed operation all the way to max DPI.

BTW, I did recommend Razer to include an "Advanced Settings" section of mouse profiles in the Razer mouse software:
- Poll Rate [500 / 1000 / 2000 / 4000 / 8000 Hz]
- Enable smoothing [On / Off]
- Enable interpolation [On / Off]
- Sensor read frequency [20KHz / 16KHz / 8KHz]

Configurable per-game.

Some games clearly work better at 2000Hz or 4000Hz, and some buggy games like Cyberpunk 2077 seems to work better at 500Hz (temporarily, until it is fixed).

Nice, hopefully these settings will be saved onboard and will be editable by something similar to Logitech's new onboard manager (I hate the bloat that comes with stuff like GHUB). I really hope this mouse will gain traction and release soon, ideally without power saving features like dynamic sensor framerate adjustments. Let us set what we want, please. Same goes for smoothing/interpolation/poll/debounce timer/LOD etc. Give customers control over their peripherals.

Futuretech wrote: ↑

24 Dec 2020, 09:46

https://www.overclock.net/threads/an-ov ... gy.1251156

I'm afraid I can't take this forum post as anything more than some sort of misinterpretation by this individual when the official USB spec says that USB 3 (both superspeed and superspeed plus) has 125us transfer intervals : https://www.intel.com/content/dam/www/p ... b-xhci.pdf

It says in multiple places that each microframe is 125us, and not a single mention of 41.6us or 24KHz anywhere in the document.

Brainlet wrote: ↑

24 Dec 2020, 23:39

Does this mouse have smoothing above 2k DPI?

he said multiple times since like a year ago there's no smoothing on anything since after 3389/90.

Chief Blur Buster wrote: ↑

24 Dec 2020, 23:44

Brainlet wrote: ↑

24 Dec 2020, 23:39

Does this mouse have smoothing above 2k DPI?

I am earlier told by Razer that this mouse is able to do support non-smoothed operation all the way to max DPI.

BTW, I did recommend Razer to include an "Advanced Settings" section of mouse profiles in the Razer mouse software:
- Poll Rate [500 / 1000 / 2000 / 4000 / 8000 Hz]
- Enable smoothing [On / Off]
- Enable interpolation [On / Off]
- Sensor read frequency [20KHz / 16KHz / 8KHz]

Configurable per-game.

Some games clearly work better at 2000Hz or 4000Hz, and some buggy games like Cyberpunk 2077 seems to work better at 500Hz (temporarily, until it is fixed).

Also, Razer_TheFriend, Cyberpunk 2077 isn't working well with the touchpad of my Razer Blade 15. Even though it worked well when I played a complete solo playthrough of Crysis 2 in a hammock during the summer on the Razer Blade 15 laptop. (You can't use a mouse in a backyard hammock during stay-at-home!)

smoothing at certain ranges or it should just be configured stock since the user doesn't know as much about capabilities and flaws as manufacturer? that is unless a power user like me who either A: owns relic/antic mouse pads that are of higher quality/standard than a lot of generic crap on the market these days, or B: actually dabbles with optimal (for various gameplay) mid to high dpi dials, where jitter of (current gen tech) sensors start to commonly get obvious, or both.

Interpolation? for 20K raw?

Razer_TheFiend wrote: ↑

25 Dec 2020, 02:51

I'm afraid I can't take this forum post as anything more than some sort of misinterpretation by this individual when the official USB spec says that USB 3 (both superspeed and superspeed plus) has 125us transfer intervals : https://www.intel.com/content/dam/www/p ... b-xhci.pdf

It says in multiple places that each microframe is 125us, and not a single mention of 41.6us or 24KHz anywhere in the document.

This is interesting, so 8000hz is the theorical maximum transfer rate of USB 3.0 since 1/0,000125 = 8000 (I guess it's the same for USB 3.1 and 3.2, correct me if I'm wrong, I'm wondering if newer USB controller can stabilize poll induced interference from high pollrate devices, this topic has been discussed a lot in this forum)

Chief Blur Buster wrote: ↑

24 Dec 2020, 23:44

BTW, I did recommend Razer to include an "Advanced Settings" section of mouse profiles in the Razer mouse software:
- Poll Rate [500 / 1000 / 2000 / 4000 / 8000 Hz]
- Enable smoothing [On / Off]
- Enable interpolation [On / Off]
- Sensor read frequency [20KHz / 16KHz / 8KHz]

Could you elaborate on Sensor Read Frequency? This is the first time I heard about it.