Northrop F-19A Specter

[Cutaway]

I know of 'F-19' models being released to the public during the Cold War to distract communist spy rings from the F-117 Nighthawk as it was being developed, But i was thinking would the "F-19A Specter" aircraft design be feasible for a Jet Fighter?

 

[AlienFTM]

Don't see why not. The configuration reminds me of the F22 Raptor now in service. It's all a question these days of whether you can pack enough computer in to fly the plane.

[Tony B]

The Northrop flying wing was in action during the 1940's. TSR2 (here we go again) was viable, and that was with hot glass electronics. Bear in mind the average MP3 player has more computer memory and power than USA used in the entire Apollo proggrame. I can remember the first commercial Intergrated Circuits.

[AlienFTM]

The Northrop flying wing was in action during the 1940's. TSR2 (here we go again) was viable, and that was with hot glass electronics. Bear in mind the average MP3 player has more computer memory and power than USA used in the entire Apollo proggrame. I can remember the first commercial Intergrated Circuits.

 

In the entire in-flight Apollo programme.

 

A lot of the behemoth mainframe as we know it today (a few days ago an IBM mainframe became the first machine ever to run in excess of 1 Petaflops - thousand trillion floating point operations per second) owes its existence to Project Apollo, notably System 360, which through System 370 and System 390 led to System z (for zero downtime. The standard service level agreement for an IBM mainframe is five minutes' unscheduled downtime per annum. One unscheduled reboot will cost you that on a Windows machine) we have at the heart of the mainframe today.

 

TSO (Time-sharing Option) owes its existence to Apollo because the massive (in size if not power compared with what is in the room next to me) mainframe needed to be able to do more than one thing at a time. It was a quantum leap in computer processing like when with Windows 95, suddenly you could have two programs actually RUNNING at once rather than one running and anything else just sitting there waiting for its turn.

 

JES (Job Entry Subsystem) was explicity written for Project Apollo. JES is an input queue where the machine sits and waits for instructions (in those days, on 80-byte punched cards: now it's card images). Throughout my mainframe career (23 years) I have been aware of a legend that if JES has an error, there are deeply-embedded error messages, never seen for decades that are prefixed NAS- (for NASA) instead of JES-. Despite my best efforts / worst incompetence, I have never managed to generate a NAS-message.

 

When input comes into the JES queue, it gets validated by an interpreter that verifies the JCL (Job Control Language) is valid; otherwise it abnormally terminates ("abends").

 

JES then starts grabbing the resources requested for the job, including data sets ("files") for input and output, hardware such as DASD ("Direct Access Storage Devices" or disks) and tape drives, processor space, etc. When all the requirements have been satisfied, JES checks for an initiator to run the job and places it in the queue for that initiator. One a modern mainframe, well-maintained by its System Programmers, there will always be enough initiators, of the right sort for the jobs coming in, just like Tesco putting the right check-out operatives on the right tills according to whether they all have large loads or small loads.

 

And the figure I was quoted for "computing power carried on an Apollo mission" was equivalent to the engine management system in a current BMW. But that was a few years ago.

 

Remember as a rule of thumb that whenever the computing world as a whole announces some stunning "new" technology, it's been there on the mainframe for 25 years.