4 ms·

The headline is completely inaccurate. The B-52H communications systems have received multiple upgrades since they were built in 1960-1961. They received a maj

by runjake 12y ago

The headline is completely inaccurate. The B-52H communications systems have received multiple upgrades since they were built in 1960-1961. They received a major revamp in the late 1980s, which is when the current CRTs and ACUs (computers) were brought in. Additionally, since their upgrade in the late 1980s, they've received numerous incremental upgrades to both the system and the software. CRTs stuck around so long, because they weren't fragile, and could handle a lot of the long term vibration and shock you deal with on an aircraft like this. Source: I worked on these systems on the B-52H for a few years. The systems even felt somewhat archaic back then, but there was a reason: they were probably the most reliable military avionics systems I've come across.

mikeash 12y ago
I'd wager (and please correct me if I'm wrong, your experience no doubt means you'd know better than I) that resilience in the face of EMP was probably a big factor too. The B-52's primary mission was nuclear bombing for a long time.
- valarauca1 12y ago
  Software doesn't deal with EMP hardening. Hardware does. Bits soft setting themselves is almost impossible to debug natively, since your debugger can be effected by the same error. Your bet met is running the algorithm ~20-100 times and comparing the outputs, then averaging. It's pretty horrible honestly.
  - arethuza 12y ago
    The US took EMP testing pretty seriously - the ATLAS-1/Trestle test rig was rather impressive: http://en.wikipedia.org/wiki/ATLAS-I http://en.wikipedia.org/wiki/ATLAS-I I found it while searching for large wooden buildings a while back: "Trestle is the world's largest structure composed entirely of wood and glue laminate"
    - valarauca1 12y ago
      Well software hardening against EMP's is pretty horrible. Hardware hardeneing is easier, just make large band gap's in your transistors.
  - mikeash 12y ago
    I'm confused. Who said anything about trying to harden against EMP in software? My understanding is that the real threat from EMP is physical destruction of the circuits by inducing voltages beyond what the wiring can handle. Any bit flips would be transient, since the EMP doesn't last very long (thus the P part). Trying to use software to protect your chips from being fried by hundreds or thousands of volts being shoved through parts that weren't designed to handle it is clearly not viable.
    - valarauca1 12y ago
      >>I worked on these systems on the B-52H for a few years. The systems even felt somewhat archaic back then, but there was a reason: they were probably the most reliable military avionics systems I've come across. >I'd wager [..] that resilience in the face of EMP was probably a big factor too The poster was talking about software reliability, which the next poster asked if software reliability was related to EMP hardening.
      - mikeash 12y ago
        It was quite clear to me that he was talking about hardware reliability, what with the explicit mentions of CRTs, fragility, and resistance to vibration and shock. The only mention of software in the whole comment is a brief comment about "upgrades to both the system and the software".
      - nathannecro 12y ago
        Systems refers to the system as a whole, not just software.
- runjake 12y ago
  Yep. Everything is nuke (EMP) hardened. Another bit of trivia: the nuclear flight crew used to have a lead eye patch, so that they'd have one good eye to operate the plane after (atomic) bomb detonation.
  - glesica 12y ago
    > so that they'd have one good eye to operate the plane after (atomic) bomb detonation. Serious question, this is quite interesting to me... why? I mean, if the Cold War had gone "Hot" there likely wouldn't have been anywhere for them to land (presumably the Soviets would have bombed the bases), they'd have been basically dead upon takeoff. Seems like they didn't even really need to carry enough fuel for a round trip, just carry enough to get to the target and fly faster. Do you know what the plan was for after the bombs went off?
    - cobrausn 12y ago
      In all seriousness, I think the plan was to make sure our bombs got dropped also. It's not MAD if you can't follow up on either the 'mutual' or the 'assured' part.
    - mikeash 12y ago
      There are some good reasons. One, a B-52 can carry more than one nuclear bomb. You want the crew to remain in flying condition after dropping the first one so they can drop more. Two, you can end up with pretty severe motivational problems if you don't at least make an effort to allow your pilots to return home. Even if you think it's unlikely that they'll make it home, you still want to give them the tools they need to try. Three, it's not entirely unrealistic that some of them would make it home. Even if every landing strip was destroyed, they could parachute out. The idea that a nuclear war would be a 15-minute orgy of MAD didn't really happen until the 70s. Before that, ICBMs were limited in number and accuracy, and bombers would be carrying out a lot of the strikes, especially on military targets. The war could take a day or two, with B-52s and other bombers making multiple roundtrips. Even afterwards, there was no guarantee that a war between the superpowers would go from zero to full-scale MAD in an instant. It wasn't out of the question that B-52s might be called on for nuclear bombing of a small number of targets in some form of limited nuclear war.

by runjake 12y ago

mikeash 12y ago

I'd wager (and please correct me if I'm wrong, your experience no doubt means you'd know better than I) that resilience in the face of EMP was probably a big factor too. The B-52's primary mission was nuclear bombing for a long time.

valarauca1 12y ago
Software doesn't deal with EMP hardening. Hardware does. Bits soft setting themselves is almost impossible to debug natively, since your debugger can be effected by the same error. Your bet met is running the algorithm ~20-100 times and comparing the outputs, then averaging. It's pretty horrible honestly.
- arethuza 12y ago
  The US took EMP testing pretty seriously - the ATLAS-1/Trestle test rig was rather impressive: http://en.wikipedia.org/wiki/ATLAS-I http://en.wikipedia.org/wiki/ATLAS-I I found it while searching for large wooden buildings a while back: "Trestle is the world's largest structure composed entirely of wood and glue laminate"
  - valarauca1 12y ago
    Well software hardening against EMP's is pretty horrible. Hardware hardeneing is easier, just make large band gap's in your transistors.
- mikeash 12y ago
  I'm confused. Who said anything about trying to harden against EMP in software? My understanding is that the real threat from EMP is physical destruction of the circuits by inducing voltages beyond what the wiring can handle. Any bit flips would be transient, since the EMP doesn't last very long (thus the P part). Trying to use software to protect your chips from being fried by hundreds or thousands of volts being shoved through parts that weren't designed to handle it is clearly not viable.
  - valarauca1 12y ago
    >>I worked on these systems on the B-52H for a few years. The systems even felt somewhat archaic back then, but there was a reason: they were probably the most reliable military avionics systems I've come across. >I'd wager [..] that resilience in the face of EMP was probably a big factor too The poster was talking about software reliability, which the next poster asked if software reliability was related to EMP hardening.
    - mikeash 12y ago
      It was quite clear to me that he was talking about hardware reliability, what with the explicit mentions of CRTs, fragility, and resistance to vibration and shock. The only mention of software in the whole comment is a brief comment about "upgrades to both the system and the software".
    - nathannecro 12y ago
      Systems refers to the system as a whole, not just software.
runjake 12y ago
Yep. Everything is nuke (EMP) hardened. Another bit of trivia: the nuclear flight crew used to have a lead eye patch, so that they'd have one good eye to operate the plane after (atomic) bomb detonation.
- glesica 12y ago
  > so that they'd have one good eye to operate the plane after (atomic) bomb detonation. Serious question, this is quite interesting to me... why? I mean, if the Cold War had gone "Hot" there likely wouldn't have been anywhere for them to land (presumably the Soviets would have bombed the bases), they'd have been basically dead upon takeoff. Seems like they didn't even really need to carry enough fuel for a round trip, just carry enough to get to the target and fly faster. Do you know what the plan was for after the bombs went off?
  - cobrausn 12y ago
    In all seriousness, I think the plan was to make sure our bombs got dropped also. It's not MAD if you can't follow up on either the 'mutual' or the 'assured' part.
  - mikeash 12y ago
    There are some good reasons. One, a B-52 can carry more than one nuclear bomb. You want the crew to remain in flying condition after dropping the first one so they can drop more. Two, you can end up with pretty severe motivational problems if you don't at least make an effort to allow your pilots to return home. Even if you think it's unlikely that they'll make it home, you still want to give them the tools they need to try. Three, it's not entirely unrealistic that some of them would make it home. Even if every landing strip was destroyed, they could parachute out. The idea that a nuclear war would be a 15-minute orgy of MAD didn't really happen until the 70s. Before that, ICBMs were limited in number and accuracy, and bombers would be carrying out a lot of the strikes, especially on military targets. The war could take a day or two, with B-52s and other bombers making multiple roundtrips. Even afterwards, there was no guarantee that a war between the superpowers would go from zero to full-scale MAD in an instant. It wasn't out of the question that B-52s might be called on for nuclear bombing of a small number of targets in some form of limited nuclear war.