No prototypes for GamesCom, but a Wifi Chip and a Pyra in the dark!


What do you think? Do you use 5GHz Wifi?
I don't, currently - but supporting 5GHz Wifi certainly is the more future-proof variant, seeing how residential areas are spammed by competing access points by now.

Also, the 2.4GHz spectrum gets polluted by lots of stuff Wifi is ill prepared for - even USB3 data transfers:

http://www.usb.org/developers/whitepapers/327216.pdf

The noise generated due to the USB 3.0 data spectrum can have an impact on radio receivers whose antenna is placed close to a USB 3.0 device and/or USB 3.0 connector. The noise is a broadband noise that cannot be filtered out, since it falls within the band of operation of the wireless device (2.4–2.5 GHz)
 
Last edited by a moderator:
GwWeWTNtX3Bw.png
Hmmmm... the Benelux really stands out on that picture. Didn't expect the difference with our neighbours to be so large.
 
I have a dell venue pro 8, When I am using 2.4ghz and BT (also 2.4ghz) I have issues with one or the other dropping out however when I use 5ghz for wifi I dont have this issue. Any chance that this will happen on the Pyra as it would be annoying to be stuck with 2.4ghz if it interferes with the BT.
 
I think that 5 GHz makes sense, too. The 2.4 GHz is really congested at my university and my flat (I live in a complex of buildings with 5 floors)


The Pyra is also capable to use the 5GHz speed boost with its faster processor
 
My experience with 5 GHz is that it is faster for large file transfers. I have switched from 2.4 to 5 GHz on the same laptop (with an a/b/g card, so actually using a rather than n) and found that sending ripped movie files (averaging around 1 GB) from the laptop to the server happened much faster with the 5 GHz connection. Generally, I have found that my devices with a 5 GHz connection and 1 or two bars on the meter were just as fast as 2.4 GHz connections with 4 or 5 bars. So I haven't really encountered much of any downside to 5 GHz. Of course, the 5 GHz devices can switch to 2.4 GHz connections if need be, so the consideration is the cost. Is the extra capability worth it? I'm not going to pretend I wouldn't like to have 5 GHz available. However, if the price is likely to really impact the probability of success for the project, then don't go for it. I'd rather have a Pyra with only 2.4GHz networking capability than not have a Pyra at all.
 
Enough people seem to want 5GHz that it is probably best to include it.  I don't personally plan to use it much, but I do think it should be there for those that need it, so I guess I am changing my preference.
 
I wouldn't mind paying extra for 5GHz, but then again, I also wouldn't mind paying extra for 3/4G. However, a lot of people have voiced their concern about the price of the Pyra.

This extra 14€ (in bare costs, so more in terms of end-user price, if only for VAT), for a mildly useful feature, seems like something that should be skipped if ED wants to make an "affordable" version of the Pyra.

Ideally, we get 3 or 4 editions of the Pyra:

  1. wifi: 2.4GHz only, no 3/4G (base price)
  2. wifi: 2.4 + 5Ghz, no 3/4G (base price + 25€ or so)
  3. wifi: 2.4Ghz only, with 3/4G (base price + 100€ or so)  --> not sure if anyone would want this
  4. wifi: 2.4 + 5Ghz, with 3/4G (base price + 125€ or so)

However, that would mean there are 3 or 4 variants of the populated PCB to create, which could introduce extra costs because of logistics etc.

Since we already have two variants of the populated PCB (one with and one without 3/4G), I would propose to make at least model 1 and 4. If having different variants is not a big deal, then we can have all 4. Or perhaps we should do a poll to get an idea of what the current community would buy? Perhaps with more accurate price estimates then, I'm just inventing numbers.

It would also be good to know what the MOQs are for these components.
 
I wouldn't mind paying extra for 5GHz, but then again, IThis extra 14€ (in bare costs, so more in terms of end-user price, if only for VAT), for a mildly useful feature, seems like something that should be skipped if ED wants to make an "affordable" version of the Pyra.
It's "mildly useful" in much the way an elevator in a 3 story apartment is. Most people may not really 'need' it on an everyday basis and could do without. Every now and then they might 'need' or be thankful to have it if they brought home groceries or a new chair or two toddlers and a pram. But - there are also potentially a few people who can't really just get by without it. In theory that elevator might be for the handicapped. In reality, it likely serves useful purpose for everyone - even if there are some who don't need it all the time.

Every now and then I sync and transfer a big pile of files to/from my Phone. For those situations, the 5 Ghz saves me -days- of transfer time.

For some people in 2.4 Ghz congested areas, the 5 Ghz means that they'll actually be -able- to use the Pyra wirelessly.

Another thing that having the 2 frequencies helps with that many of you might not consider:

Pyra as a network bridge. If you need me to explain that, then it likely doesn't apply to you.

Please - the 5 Ghz may seem like 'extra' until you actually need it, but when you need it - you really do need it.
 
Still with 2 and 4. There's no need for an extra pay wall just to get 5Ghz.
 
I'm just curious - does this 5GHz extension to wifi chip adds also support for 802.11ac ? If yes, then the price might be justified.

I was also hoping that 3/4 G won't as expensive. I was expecting 30 - 50 €.
 
Multiple hardware configurations is huge PITA from OS maintainer's point of view though.
In general, yes. In this case though: wouldn't the two variants of the wifi chip use the same kernel module, which somehow sees the chip capabilities at initialization? Or perhaps two kernel modules, where the correct one gets loaded automatically based on some identification code or something. I'm not a kernel hacker, but that's the kind of thing I would expect. And the higher-level networking stuff probably doesn't care at all about the underlying wifi protocol, it sees abstractions of that anyway. So I don't really see where the huge PITA would be in this case.

Debian was designed to run on a huge variety of hardware configurations, so I don't think it would be a particularly big problem to have a few variants of the hardware.

An affordable "base model" and a more expensive "premium model" seems like a good model to me, and something like that was the plan all along w.r.t. the cell modem for 3/4G and GPS. "Intermediate" models are probably going to complicate matters (mostly from the logistics point of view to source the components in the right proportions), but two models sounds like a good plan, and then I would do everything to make the base model as cheap as possible (perhaps even go as far as giving it a smaller eMMC and less RAM) and the premium model as perfect as possible.
 
I'm just curious - does this 5GHz extension to wifi chip adds also support for 802.11ac ? If yes, then the price might be justified.

I was also hoping that 3/4 G won't as expensive. I was expecting 30 - 50 €.
I don't think so. Marketing would have written AC and crazy bandwidths all over if it were the case. 5Ghz would be N speed, 1X1 so 150Mbit/s theoretical.
 
I'm just curious - does this 5GHz extension to wifi chip adds also support for 802.11ac ? If yes, then the price might be justified.

I was also hoping that 3/4 G won't as expensive. I was expecting 30 - 50 €.
I don't think so. Marketing would have written AC and crazy bandwidths all over if it were the case. 5Ghz would be N speed, 1X1 so 150Mbit/s theoretical.
From what I can tell, no. From what I can tell, the dual band module would be the WL1837MOD, which only seem to support abgn.
 
Back
Top