This is merely a historical archive of years 2008-2021, before the migration to mailman3.
A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/UmTRX@lists.osmocom.org/.
Jean-Samuel Najnudel - BJT PARTNERS SARL jsn at bjtpartners.comHi Alexander, I fully agree with your vision. Especially that "the best is an enemy to a good". Moreover, it is also very important for me to deploy the UmTRX in Mayotte ASAP as my local partners would pressure me quite a lot in the near future. However, I really belive this preselector is not just for scoring the best performances as possible as an RF geek, cool and fun challenge. We really need a preselector, not only to pass the GSM specs. I do not care that much about the GSM specs, especially the hard to pass macro BTS specs. However, as I mentioned you in Berlin in last december, I really think sampling the whole 1.5 MHz band would be a real problem. The preselector is good to pass the spec but it is also the only reasonable solution to lower the sampling band down to about 200 KHz. Actually, I would really need this. Even for my experiment network in Mayotte, this would not be realistic to deploy the network and provide reliable services without this preselector. I really think this preselector is mandatory, in practice, even if we do not care about the GSM specs. Without this, I would not even be able to deploy the UmTRX in Mayotte. I agree with you when you mention some markets would still not need this preselector. First market is lab equipments. However, I am not sure volumes would be that high, especially as even a basic low cost SSRP is good enough for many lab use cases. Second market is the femtocell market. However, for this market, the UmTRX would anyway be too expensive for most business cases. Except ip.access who can sell much more than everybody else because of their first mover advantage, most of the femtocell vendors target list prices around 300 Euros (eg: HSL) and even lower (most 3G femto vendors). Considering this competition, I believe it would be really difficult to succeed in the femtocell market with the current UmTRX design. We would need something even more low cost orienteed and probably also more embedded. The market I relly believe in is the mid-range mid-capacity BTS market. There are many high performance macro BTS vendors. There are many simple, cheap and reliable femtocell vendors. There are not much mid-range BTS vendors. If we want something than can cover a few kilometers range, that is energy efficient and which use IP standards, you only have something like 3 vendors: Vanu, VNL and IP.access. These guys are really expensive. Offering a better value product would not be that difficult, at least compared to the other markets. Moreover, this mid-range mid-capacity BTS market is potentially huge. This is the most suitable system we would need to cover most of the next 2.5 billion guys who do not have a reliable and affordable phone service (sometimes just no service at all). Again, I agree with you, Time To Market is very important. However, I do not think a few weeks delay would make you miss the opportunity. Also, the customers on this market (mainly rural and local carriers) would have very similar needs as mine. Most of them would probably need to get a narrow sampling band (< 1.5 MHz) to avoid problems with interferences (1st TRX to the other, other existing carrier...) causing, among other issues, ADC saturation. As me, these customers would need the preselector. The preselector would anyway need a VCO/PLL, a mixer and an IF SAW filter, even if we do not expect to pass the GSM spec. Why not trying to select good performance components to try (not guarantee but just try) to pass this spec. This would not need much more research and development efforts and total BOM will not be that much more expensive. Moreover, we can really expect some of your customers would not agree to deploy a system that does not pass the GSM spec. A mid-range BTS would need a Tx power around 5 to 10 Watts. Above 2 Watts, the hardware is considered as a macro BTS. Considering this, I really think we should try (not take one year delay for this but just try) to pass the GSM macro BTS spec. Honestly, I may be wrong but I do not believe Andrey will take that much time to design the preselector. For the BOM extra cost, we talk about 60 USD per UmTRX board. I agree with you, this is not nothing but this really worth the money. I actually believe value would be really good with this. Even with my very tight budget, this would not be able problem for me to pay this. The extra cost would be much lower than all the worries and time spent on frequency planning, interferences and blockers problem solving on the field. Anyway, I also really think, even with the high performance preselector, the UmTRX system is still really cost effective compared to the competitors. I really feel confident about this. Best regards. Jean-Samuel. :-) On Sun, Mar 18, 2012 at 8:45 AM, Alexander Chemeris < alexander.chemeris at gmail.com> wrote: > Andrey and all, > > 18.03.2012 2:31 пользователь "Andrey Sviyazov" <andreysviyaz at gmail.com> > написал: > >> Regarding the GSM spec, I believe these blocker tests are hard to pass > and not that useful in most practical situations. However, I do not care > that much about passing this spec for my network deployment in Mayotte but > I really believe passing the spec will be very important for you if you > wish to sell your hardware solution to some major operators. Moreover, as > we would anyway need a superheterodyne selective filtering to get a > reasonably narrow Rx sampling band (< 1.5 MHz LMS band), it does not cost > that much more to try to pass the GSM spec. > > > > You right, it is important for us and may be it is really important for > systems with high channel dencity. > > I completely support everything which could make our product better > without increasing its cost. But we also must ensure to release UmTRX in > time. What we all should keep on mind is that "the best is an enemy to a > good". We should focus on those 10% of simple tweaks which bring us 90% of > improvement. Otherwise we'll be swamped with the other 90% of tweaks and > will miss the market opportunity. I can't stress it more - we MUST release > UmTRX ASAP. Even if it doesn't meet macro-BTS requirements. We'll be able > to fix this in the next version if ever needed - we can't know the real > demand until we release the first version. > > I would be glad of what I've just said is obvious and already lives in > your heart. Otherwise it's extremely important you understand this deeply, > not formally. Please let me know if you don't, I'll explain in more details. > > > Furthermore, it is very easy to lose the reputation of the product, but > it is very difficult then to fix it back. > > This is true. And the best way to keep the reputation is to realistically > understand UmTRX capabilities and avoid overmarketing. In other words, with > just reasonable product quality, our reputation depends solely on the right > marketing. E.g. we might explicitly warn customers that it's not suitable > for macro-BTS installations and they could do so on their own risk only. > > > I think, low cost doesn't sign low quality, so we must to have good > hardware on market for good sales and promote OpenHW :) > > This is very true. Just keep in mind that "good hardware" means "minimally > viable hardware at low price" and doesn't mean "super high quality mumbo > jumbo with many zeros in the price". Our customers value simplicity and low > cost over complexity and golden plates. > > That said, I can't help with decisions on the RF side and here I rely on > you, guys. That's why it is so important for you to understand all these > "abstract" marketing ideas. > > Sent from my Android device. > > -- > Regards, > Alexander Chemeris > CEO, Fairwaves LLC > http://fairwaves.ru > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.osmocom.org/pipermail/umtrx/attachments/20120318/80358b89/attachment.htm>