For those who are worrying about the longevity of Allwinner V3s

[technix]

This is from my visit to Allwinner stall at CES Asia yesterday. I grabbed an Allwinner rep and inquired about the long term availability of V3s, the BGA-free SBC chip.

"Allwinner can guarantee at least five years for the availability of V3s, and the same die is also being used in a few other parts for those that needs second sources or replacement parts."

[ebastler]

"the same die is also being used in a few other parts for those that needs second sources or replacement parts."

Does that mean "used in other AllWinner parts" or in third-party parts?

If it's the former, that would not really qualify as a second source, right? If it's the latter, I would love to know which are the other brands. And also would love to know whether it is common for multiple brands to share dies?

Thanks for further background, if you know more!

[chickenHeadKnob]

That probably would be me! 

I am no longer an interested in them, as I can slowly accumulate PI zero and zero/W enough for hobby projects. In particular I can get Pi zero/W  with pre-soldered header in unlimited  (x) quantity from an official Canadian Pi distributor BuyaPi.ca for $19 cad ($14.26 usd), At that price it is competitive with orange PI and similar. I suspect Pi foundation longevity of supply is even more trustworthy than whatever Allwinner tells you. With the official Pi I don't worry about the  BGA soldering and it has more RAM.

[nctnico]

This is from my visit to Allwinner stall at CES Asia yesterday. I grabbed an Allwinner rep and inquired about the long term availability of V3s, the BGA-free SBC chip.

"Allwinner can guarantee at least five years for the availability of V3s, and the same die is also being used in a few other parts for those that needs second sources or replacement parts."

You take a salesperson's word for something that important?    If you depend on long term availability (and 5 years isn't long term at all) you better get something in writing. NXP (for example) has guaranteed availability of ARM SoCs for up to 15 years (from introduction). Now that is something you can base an industrial design on.

[rsjsouza]

This is from my visit to Allwinner stall at CES Asia yesterday. I grabbed an Allwinner rep and inquired about the long term availability of V3s, the BGA-free SBC chip.

"Allwinner can guarantee at least five years for the availability of V3s, and the same die is also being used in a few other parts for those that needs second sources or replacement parts."

You take a salesperson's word for something that important?    If you depend on long term availability (and 5 years isn't long term at all) you better get something in writing. NXP (for example) has guaranteed availability of ARM SoCs for up to 15 years (from introduction). Now that is something you can base an industrial design on.

I was going to reply the same thing back then, but the silence of the OP about ebastler's questions made me dismiss the assertive as just vapourware.

Five years is enough for a production run of the newest cheap gadget that will ve replaced next year with a new thing.

[technix]

You take a salesperson's word for something that important?    If you depend on long term availability (and 5 years isn't long term at all) you better get something in writing. NXP (for example) has guaranteed availability of ARM SoCs for up to 15 years (from introduction). Now that is something you can base an industrial design on.

I am not in a position that can ask them for anything in writing. In fact should it not be CES they would not even bother answer my inquiries. Allwinner HQ has a general refusal of response to any and all low volume customers.

Five years is enough for a production run of the newest cheap gadget that will ve replaced next year with a new thing.

V3s being now sold as an automotive market chip tells me something differently.

[chickenHeadKnob]

There are two posts on the Olimex blog that explain AllWinners  policy for discontinued chips as Olimex still supplies A13 boards and others to the industrial market.
see may 13 post in https://olimex.wordpress.com/2019/05/16/long-term-olinuxino-supply-how-it-works/

and https://olimex.wordpress.com/2014/11/27/how-long-olinuxino-with-allwinner-socs-will-be-produced-again-now-we-know-the-answer-forever/

[splin]

There are two posts on the Olimex blog that explain AllWinners  policy for discontinued chips as Olimex still supplies A13 boards and others to the industrial market.

So it's all hunky dory providing you can order 50K pieces at a time. Or you buy boards from Olimex and they have enough customers to justify buying 50k pieces periodically - but how long might you have to wait before Olimex have received enough orders/requests to justify buying another 50k pieces? At some point declining orders means they won't re-order  50k parts just because their stock drops below the usual re-order threshold if their sales projections can't justify it.

Not easy is it?
 

[rsjsouza]

Funny how the poorly written blogs try to convey the idea of supply chain reliability and endurance when the company itself only exists for a fraction of what they promise. This is not gained by blog posts but by track record, which takes a long time to build.

The (well deserved) criticism of western companies changing ownership all the time is eroded by the lack of a balanced view that many companies in China simply ceased to exist or changed names over all these years.

To me these are non independent ad posts - take their info with a *pound* of salt.

[chickenHeadKnob]

There are two posts on the Olimex blog that explain AllWinners  policy for discontinued chips as Olimex still supplies A13 boards and others to the industrial market.

So it's all hunky dory providing you can order 50K pieces at a time. Or you buy boards from Olimex and they have enough customers to justify buying 50k pieces periodically - but how long might you have to wait before Olimex have received enough orders/requests to justify buying another 50k pieces? At some point declining orders means they won't re-order  50k parts just because their stock drops below the usual re-order threshold if their sales projections can't justify it.

Not easy is it?

Yes I came to that same conclusion  years ago. I am not an Olimex customer, but my intuition tells me they  tend to the industrial/kiosk end with some additional hobby market and they have been pestering OLIMEX about the reliability of supply. So I mentioned those blog posts because I thought they were topical to this thread and I didn't want to editorialize them.  AllWinner is fabless  (I think) and the report that they would revive an old part from their catalogue if you throw enough money at them rings true.

Here is a post I made in 2015, I'll quote in full:
 
@Stigaard

For me the comparison wouldn't be a broadcom or Rockchip part, they might be as bad or worse, I would compare to Freescale i.Mx or TI. soc. that are intended for the embedded industrial control market and have some promise of continuity of supply and good documentation. In TI case they claim 10 year availability or similar. I suppose I am risk averse and I view time spent designing in and committing to a new chip as a big risk. If you don't mind throwing that  effort away you may feel differently.

 Olimex is selling Allwinner boards into this type of market and their customers often ask, "how long will this be available". Tsevtan recently replied "don't worry Allwinner is stopping production of A10 but as long as we order 50,000 we can get them forever into the future".  Well I would worry, I was with a small manufacturer and we got put on allocation a few times. That was pain I will long remember. That promise from Olimex is unrealistic for two reasons. First the mask is tied to a particular scale  "node"  and when the fab changes, and they change on their own schedule not Olimex's whim, then that's it. In addition there is the negative feedback effect, Olimex tells customers wait until we get to 50,000 ordered and the customer will say we can't wait and then designs out the part. That will mean it never gets to 50000.

From the thread -
https://www.eevblog.com/forum/microcontrollers/mediatek-mt5389/msg580840/#msg580840
edit: mucked up link

[tszaboo]

Five years is enough for a production run of the newest cheap gadget that will ve replaced next year with a new thing.

V3s being now sold as an automotive market chip tells me something differently.

Chinese car manufacturing is not considered automotive, it is consumer.
We will talk about cars again, when one of them accidentally they got their first star on an European safety test.

[tszaboo]

Chinese car manufacturing is not considered automotive, it is consumer.

Every Chinese car model has to pass Chinese government's compulsory certification process, which the standard is directly derived from European ISO standards.
Automotive electronics giant Infineon made $2.6BN in greater China region, more than all other markets, including EMEA. Guess where those parts went?

We will talk about cars again, when one of them accidentally they got their first star on an European safety test.

That's a stupid, narrow sighted and sinophobic statement. Chinese car companies don't sell in EU, so why would they get Euro NCAP ratings in the first place?
Rest of 100% Chinese designed EVs, China doesn't export cars at all due to IP restrictions (most Chinese cars are built on Western IPs, which part of the agreement is no market invasion).

Buit they were tested in India to the Euro NCAP. And south america. They dont even have ESS.
https://dailykanban.com/2014/11/double-standards-5-star-rated-chinese-cars-failed-euro-ncap-rules/

BTW Germans have tested the "Brilliance" BS4 and BS6, awarding both 0 stars.
And your arguments are only strengthening my points. And nobody really knowing me would accuse me of being sinophobic.

[tszaboo]

Based on the lacking of ESP?

No, based on "nothing will survive in this piece

https://www.adac.de/_ext/itr/tests/Crashtest/CT0222.pdf
This is the kind of bullshit you get from russian paper cars, and USA muscle cars designed by farmers (big engine, mo powa).
Anyway, we are off topc.

[rsjsouza]

(just to add to the off-topic, but just a bit). In Brasil, brands like Chery, Lifan or Geely sell their cars there, but they are considered very poor quality by the specialized press. And that is saying a lot in a country where the quality of the big brands is considered tremendously low by worldwide standards.

Never having driven one, I can't tell further.

[technix]

BTW Germans have tested the "Brilliance" BS4 and BS6, awarding both 0 stars.

Have they tested something higher end like FAW Hongqi series? Brilliance is a low-end brand and those two are low-end model to begin with. There is no point complaining a Prius being less reliable than a Tesla Model X.

[ebastler]

There is no point complaining a Prius being less reliable than a Tesla Model X.

Are they? There is a big difference between omitting fancy features, top engine power or torque on the one hand, and compromising safety or reliability on the other hand. I don't think Toyota compromises the latter, or could get away with doing that -- despite the fact that they are not "luxury cars" like the Tesla Model X.

That may be a key difference to China: It appears that Chinese entry-level car brands can get away with compromised safety?

[ebastler]

That may be a key difference to China: It appears that Chinese entry-level car brands can get away with compromised safety?

The so called safety features were not mandated 10 years ago, and people have been driving for over a century.

The discussion was not about ESP, but about crash tests -- mechanical safety by design. Which protects drivers and their passengers not only from the results of their own driving, but also from third parties' behavior, which is out of their control.

To my (Western) mind, building cars which sacrifice solid safety design is just like allowing unsafe work practices or massive environmental pollution in the production of consumer goods: For the sake of profit, and maybe the sake of making cheap toys available to consumers, the manufacturer puts human lifes and health at risk.

[NiHaoMike]

The discussion was not about ESP, but about crash tests -- mechanical safety by design. Which protects drivers and their passengers not only from the results of their own driving, but also from third parties' behavior, which is out of their control.

To my (Western) mind, building cars which sacrifice solid safety design is just like allowing unsafe work practices or massive environmental pollution in the production of consumer goods: For the sake of profit, and maybe the sake of making cheap toys available to consumers, the manufacturer puts human lifes and health at risk.

What about motorcycles with even less safety?

I actually think that drivers with a history of behavior (e.g. drunk or reckless driving) that put the public at risk should be restricted to driving only motorcycles or flimsy cars, in order to push the risk towards them and away from the innocent public.

[ebastler]

What about motorcycles with even less safety?

Yes? What about them?
https://en.wikipedia.org/wiki/Whataboutism

[tszaboo]

Since discussion is completely OT, mind as well...

The discussion was not about ESP, but about crash tests -- mechanical safety by design. Which protects drivers and their passengers not only from the results of their own driving, but also from third parties' behavior, which is out of their control.

To my (Western) mind, building cars which sacrifice solid safety design is just like allowing unsafe work practices or massive environmental pollution in the production of consumer goods: For the sake of profit, and maybe the sake of making cheap toys available to consumers, the manufacturer puts human lifes and health at risk.

What about motorcycles with even less safety?

I actually think that drivers with a history of behavior (e.g. drunk or reckless driving) that put the public at risk should be restricted to driving only motorcycles or flimsy cars, in order to push the risk towards them and away from the innocent public.

Well, right now, you are 50 times more likely to die if you ride a motorcycle or a moped. In fact, most people think motorcyclists should be organ donors by default, without their consent.

That may be a key difference to China: It appears that Chinese entry-level car brands can get away with compromised safety?

The so called safety features were not mandated 10 years ago, and people have been driving for over a century.

Yes, and the number of fatal crashes has been steadily decreasing.

The cyclists are up there, cause they refuse to wear a helmet, and their eyes are glued to phones.
And these are pretty much the safest roads on the planet.

BTW Germans have tested the "Brilliance" BS4 and BS6, awarding both 0 stars.

Have they tested something higher end like FAW Hongqi series? Brilliance is a low-end brand and those two are low-end model to begin with. There is no point complaining a Prius being less reliable than a Tesla Model X.

Lol why would they, they are not even considered to be sold there. I'm quite sure you think of these cars as the pinnacle of technology, but a European will look at it, and call it hideous. And they dont crash test all cars, for example you will not see a Catherham crash test or a Rolls-Royce. Cause limited money, buyer not interested in crash test result and all that.

And there is a point comparing them. Seriously, dont go into an argument about things that you just assume, and easy to disprove.
https://www.forbes.com/sites/jimgorzelany/2018/10/25/the-most-and-least-reliable-rides-on-the-road/#6e8b5a0652c7
So apparently, Teslas are shoddy work, and Toyotas are reliable.
"While the brand’s latest EV, the Model 3, is predicted to deliver average reliability, that’s not the case with the pricier Model S sedan and Model X SUV. The latter has long been ranked poorly for reliability by CR, with owners most perturbed by issues with the vehicle’s falcon doors and its oversized dashboard-display screen. The Model S has been up and down over its lifespan with regard to CR’s ratings. It’s been downgraded to a “below average” rating this year, due in part to suspension problems and other concerns that include the car’s finicky extending door handles."
Wow, they already achieved average reliability.

[NiHaoMike]

Well, right now, you are 50 times more likely to die if you ride a motorcycle or a moped. In fact, most people think motorcyclists should be organ donors by default, without their consent.

If they ban cheap, flimsy cars for being unsafe, why don't they also ban motorcycles for being even more unsafe? I think an analysis of the (reduced) risks such vehicles pose to others would help support the idea of restricting bad drivers to those, increasing the safety of the good drivers.

[technix]

(just to add to the off-topic, but just a bit). In Brasil, brands like Chery, Lifan or Geely sell their cars there, but they are considered very poor quality by the specialized press. And that is saying a lot in a country where the quality of the big brands is considered tremendously low by worldwide standards.

They have a different target market than you. Those cheap cars are intended to be driven by folks in rural locations in China so they can get their produce out to the marketplace. Rural places in China can be very hilly and often lacked public transport infrastructure. The point of those cars isn't to be the safest thing to drive in, but to be cheap enough for those rural folks so they can earn some income to upgrade later. Those cars are even subsidized by the government for this purpose.

[technix]

I actually think that drivers with a history of behavior (e.g. drunk or reckless driving) that put the public at risk should be restricted to driving only motorcycles or flimsy cars, in order to push the risk towards them and away from the innocent public.

Or how about life ban on driving and prison time of no less than 6 months? In China there is an official felony called "Reckless Driving", which covers all DUI and reckless driving cases, and it comes with a minimum prison time of 6 months. A convicted felon of Reckless Driving is banned from driving anything for life, as well as various other lifetime restrictions and social stigma of being a convicted felon.

The discussion was not about ESP, but about crash tests -- mechanical safety by design. Which protects drivers and their passengers not only from the results of their own driving, but also from third parties' behavior, which is out of their control.

To my (Western) mind, building cars which sacrifice solid safety design is just like allowing unsafe work practices or massive environmental pollution in the production of consumer goods: For the sake of profit, and maybe the sake of making cheap toys available to consumers, the manufacturer puts human lifes and health at risk.

AFAIK China requires all car designs to pass a crashworthiness test before it can enter mass production, however this test is carried out with China CCC standard, a different standard from Euro NCAP. Even car technology research have to base on a CCC-certified platform for it to be legal to have test runs on public roads. (Shanghai Tongji University College of Automobiles have a fleet of modified SAIC-VW cars as their test platforms. They have free access to SAIC test track, however only ones modified from CCC-certified mass market cars without compromising its safety features are permitted to have test runs on public roads, especially long distance tests along national expressway G2 to and from Beijing)

[tszaboo]

Well, right now, you are 50 times more likely to die if you ride a motorcycle or a moped. In fact, most people think motorcyclists should be organ donors by default, without their consent.

If they ban cheap, flimsy cars for being unsafe, why don't they also ban motorcycles for being even more unsafe? I think an analysis of the (reduced) risks such vehicles pose to others would help support the idea of restricting bad drivers to those, increasing the safety of the good drivers.

They made several safety equipment mandatory on bikes. For example ABS, rear view mirrors etc.
Also, they are setting up new environmental standards for motorcycle emissions. This has the indirect effect, that big engined cycles are now dont sell in the EU. What happens is that manufacturer X has a sales of 1000 cycles in across the entire EU from a certain model, and the complying with the emissions would cost them for example 3 million EUR. And they cannot increase the price by 3000 EUR (including the extra equipment), cause nobody would buy the cycle, and therefore they just abandon the market. Good riddance. BTW they are not allowed to ride on the cycle ways, have stricter speed limits etc... So they do things to limit it. But they cannot just ban it.

If you would put bungee jumping on the list, probably there is more bungee jumping accident per million KM than driving, but you are not going to ban that, do you.

[ebastler]

The cyclists are up there, cause they refuse to wear a helmet, and their eyes are glued to phones.

And because the authors chose to show accidents per distance driven, and bicyclists and pedestrians are slow. If you were to plot "accidents per hour spent in traffic", the data would look somewhat different.