R211 Discussion Thread

[zacster]

On 6/13/2023 at 4:20 PM, CenSin said:

I mean… that’s still consistent with what I said. The solution (software, hardware, configuration) remains the same, but the data that it processes (i.e., the audio files) changes.

All it takes is one weak link in the chain requiring backwards compatibility to spoil the potential tech advancement options for the rest. For example, the original SD standard capped memory cards to 2 GB. If some hardware were developed around a platform that hadn’t updated its controllers, then every future product developed around that platform would be held back on SD card capacity.

You have that wrong.  2GB is the maximum address space of a 32 bit number.  +/- 2^31.  It was forced on them by the math.  That was a limit in a lot of systems, in fact all 32 bit systems.  That's one reason they went to 64.  They had all this capacity that couldn't be addressed directly.

A funny story:  I was the administrator of the financial consolidation system of a major corporation.  One day I was invited to a meeting with the Finance people (it wasn't all that rare as I knew more than they did most of the time.) They were having this problem with a small program that they had a contractor write for them that compared the financial consolidation system numbers with what was going to be published to the SEC.  It was sourced from the consolidation but this was a final check to make sure they'd captured all the latest updates,  The problem they had was they kept getting false negatives, the numbers didn't compare when they were the same.  They pulled up an example and I noticed something so I went to Excel while they talked and it confirmed what I thought.  They had exceeded the precision of a 64 bit number and the systems used approximations and these didn't match!  I of course chimed in and gave them this result and they looked at me like I had 3 heads (one more than they usually thought I had.)  The numbers were in the 100s of Billions (yep, that's right Billions with a B)  and out to 6 decimal places.  18 places of precision were needed and it exceeded the capacity of Excel.  I rewrote the entire system for SEC reporting and that never was an issue again.

I digress though, back to the R211.  I got to ride it once entirely on luck.  I never take the A but the one time I did there it was.

And didn't Bill Gates once say 640kb is all we'll ever need?  (maybe not)

Edited June 15, 2023 by zacster

[CenSin]

1 hour ago, zacster said:

You have that wrong.  2GB is the maximum address space of a 32 bit number.  +/- 2^31.

I’m intimately familiar with the limitations as I have been using SecureDigital cards since the initial 128 MB capacity. The 4 GB SDHC cards required a new card reader because the larger capacity was not backwards compatible with older readers. Same with SDXC when that came out. And so far, no SecureDigital card on the market exceeds 1.5 TB (of which Micron is the sole manufacturer).

When I say the “original SD standard,” I mean to the exclusion of SDHC (SecureDigital High Capacity), SDXC (SecureDigital Extended Capacity), and SDUC (SecureDigital Ultra Capacity). The limits of each type are listed below (c.f., https://www.sdcard.org/consumers/about-sd-memory-card-choices/sd-sdhc-sdxc-and-sduc-card-capacity-choices/):

• SD: 2 GB
• SDHC: 32 GB
• SDXC: 2 TB
• SDUC: 128 TB

This is without the additional complication introduced by physical card sizes, supported protocols, bus speeds, minimum bandwidth, or minimum IOps.

[ABCDEFGJLMNQRSSSWZ]

Really keeping my fingers crossed that the pilot will pass the 30 day test and the other R211s will begin to enter service. I will miss the additional seating of the R46s, but it's kind of perfect timing because I won't be relying on the  as much now that HS is over.

I don't think seating should be too bad on an issue on the  though; I've almost always been able to get a seat on the 8-car R179 sets regardless of if it's rush hour, the weekends, or another time. A 10-car set of R211s would be a net improvement in total seating capacity.

Also in practice, I feel like the R46's seating capacity is a bit overrated because yes, while an 8 car R46 train has physically more seating than a 10 car NTT, in practice the window seats often become blocked off and people seem to be more willing to have their backpacks or whatever take up a seat as well on the R46s.

[AZthefoamer]

Today should be the 27th day of testing if my calculations are correct. If all goes well it will end on Sunday which could mean that the next 2 trains will be in service next week.

[Ale188]

2 hours ago, AZthefoamer said:

Today should be the 27th day of testing if my calculations are correct. If all goes well it will end on Sunday which could mean that the next 2 trains will be in service next week.

Yep. You are correct. It’s kinda interesting its 30th day  of testing is on Father’s Day. I have 6/19 on my calendar rn cuz that testing will be done by then if no issues get caused.

[Chris89292]

3 hours ago, AZthefoamer said:

Today should be the 27th day of testing if my calculations are correct. If all goes well it will end on Sunday which could mean that the next 2 trains will be in service next week.

We’ll see soon enough, hopefully the MTA doesn’t find any “propulsion issues” that takes about a month to repair apparently 

[Comrade96]

22 hours ago, Chris89292 said:

We’ll see soon enough, hopefully the MTA doesn’t find any “propulsion issues” that takes about a month to repair apparently 

there were other issues wrong with the set besides the propulsion

[MJHmarc]

1 hour ago, Comrade96 said:

there were other issues wrong with the set besides the propulsion

Generally all software related. 

[AZthefoamer]

Is it concerning that (as far as I know) there have not been any R211A deliveries in over 2 months?

[subwaycommuter1983]

1 hour ago, AZthefoamer said:

Is it concerning that (as far as I know) there have not been any R211A deliveries in over 2 months?

Not really.

There are production trains (about 2-3 ten car trains) on MTA property and the pilot r211S train plus the second half of the soft shell open gangway train was  delivered within the last few weeks. 

Don't be surprised if we see more production trains being delivered next week.

[jon2305]

The R211A's current 30 day revenue test is close to the finish line. If or when all goes well at least within the next few days, then things would get rollin' soon enough.

[Around the Horn]

3 hours ago, AZthefoamer said:

Is it concerning that (as far as I know) there have not been any R211A deliveries in over 2 months?

They stopped for a while because the track they used to enter the system was under construction. They accepted another R211T set last week so we should be good now for further deliveries

[darkstar8983]

8 hours ago, Around the Horn said:

They stopped for a while because the track they used to enter the system was under construction. They accepted another R211T set last week so we should be good now for further deliveries

Mmm that should mean that for the short term we will be seeing regular deliveries since in that interim more cars should have been theoretically completed at Kawasaki

[Fan Railer]

10 hours ago, jon2305 said:

The R211A's current 30 day revenue test is close to the finish line. If or when all goes well at least within the next few days, then things would get rollin' soon enough.

that's under the assumption there wasn't another clock reset that we don't know about lol.

[RedLine]

(delete)

Edited June 17, 2023 by RedLine

[JustTheSIR]

3 hours ago, Fan Railer said:

that's under the assumption there wasn't another clock reset that we don't know about lol.

Imagine on the last day just before it reaches the terminal on the last run it has another propulsion problem

[subwaycommuter1983]

7 minutes ago, JustTheSIR said:

Imagine on the last day just before it reaches the terminal on the last run it has another propulsion problem

Let's be optimistic. 💪💪

[zacster]

On 6/14/2023 at 11:59 PM, CenSin said:

I’m intimately familiar with the limitations as I have been using SecureDigital cards since the initial 128 MB capacity. The 4 GB SDHC cards required a new card reader because the larger capacity was not backwards compatible with older readers. Same with SDXC when that came out. And so far, no SecureDigital card on the market exceeds 1.5 TB (of which Micron is the sole manufacturer).

When I say the “original SD standard,” I mean to the exclusion of SDHC (SecureDigital High Capacity), SDXC (SecureDigital Extended Capacity), and SDUC (SecureDigital Ultra Capacity). The limits of each type are listed below (c.f., https://www.sdcard.org/consumers/about-sd-memory-card-choices/sd-sdhc-sdxc-and-sduc-card-capacity-choices/):

• SD: 2 GB
• SDHC: 32 GB
• SDXC: 2 TB
• SDUC: 128 TB

This is without the additional complication introduced by physical card sizes, supported protocols, bus speeds, minimum bandwidth, or minimum IOps.

But you are still not understanding that the 32 bit word is the limitation.  Everything after that needed special programming to make it work in a 32 bit OS.  I'm not saying that's not possible, it certainly was done,  just that 2gb is not a random limitation.  We had a system that could go to 4gb by turning a switch.  It used the sign bit, but the program recognized that.  The entire system went through an upgrade to 64 bit OS and problem solved. 

[CenSin]

36 minutes ago, zacster said:

But you are still not understanding that the 32 bit word is the limitation.  Everything after that needed special programming to make it work in a 32 bit OS.  I'm not saying that's not possible, it certainly was done,  just that 2gb is not a random limitation.  We had a system that could go to 4gb by turning a switch.  It used the sign bit, but the program recognized that.  The entire system went through an upgrade to 64 bit OS and problem solved. 

You’re missing the forest for the trees here. The original SecureDigital card limitations weren’t OS-level limitations. They were baked into the standard which had to be revised to allow for a bump in maximum capacity. Perhaps the 2 GB limit and 32-bit addressing coincide, but if that were the case, how do you explain the 32 GB limitation of SDHC? Or the 2 TB limitation of SDXC? Or the 128 TB limitation of SDUC? Are SDHC cards using 35-bit microcontrollers?

[Vulturious]

These are not my photos obviously so credit to original owners. I'm not surprised the R211S has NYCT programming, it was also signed up as the  to Broad St.

[frogiefish]

I’m planning to catch the first trip on the R211A out of 207th Street tomorrow. Does anyone have a rough schedule or timetable for the R211A?

[RedLine]

If there is one, there are a few reroutes this weekend that will make it moot.

[trainfan22]

26 minutes ago, frogiefish said:

I’m planning to catch the first trip on the R211A out of 207th Street tomorrow. Does anyone have a rough schedule or timetable for the R211A?

207th is closed this weekend due to a G.O, is starting/ending at 168th.

Edited June 17, 2023 by trainfan22

[frogiefish]

Oh right

Edited June 18, 2023 by frogiefish

[1998NewFlyer]

Last day of the 30 day test 🤞🏻

Edited June 18, 2023 by 1998NewFlyer