I'm trying to do some homework on anti-skip technology in 90s/2000s portable CD players. Sony had technology called ESP, which looks like it did lossy compression and offered 10 seconds of anti-skip protection. Then I see Sony created ESP2, which says it uses a DSP, and not buffer memory. Not sure exactly what this means. I also have a Panasonic SL-CT582v, which has buffer memory and has 2 different anti-skip settings. Th 10 second anti-skip uses lossless compression, and 45 second anti-skip uses lossy compression. Are there any portable CD players with a decent amount of lossless anti-skip?
Anti-Shock and bitstrem CD players appeared at around the same time. We had a bit of a debate about both technologies here http://stereo2go.com/forums/threads/bottom-fishing-for-cd-players.6390/ Feel free to add any more information you find.
The key question is as to what memory devices were used. Yesterday I picked up a Sony D-E226CK Walkman CD Player with what they call ESP Max. Maybe you can find out where that fits in the different schemes. Unfortunately leaking batteries have destroyed the board. To answer your question and my curiosity I downloaded the service manual from Elektrotanya. The memory used in it is this one https://pdf1.alldatasheet.com/datasheet-pdf/view/11188/OKI/MSM51V4400.html 1MByte x 4 = 512KBytes A CD bitstream is 44,1KHz x 2 channels x 16 bits = 176KBytes/ second. So without some kind of compression you would get about 3 seconds buffering.
What is a decent amount? I have a Sony player with 10 s ESP that can be turned off. I also have a SL-582V like yours. I think 10 s is enough unless you use it for jogging. There were some late Philips models with more than 200 s ESP. "1MByte x 4 = 512KBytes" -- this some weird math. 1024 Ki == 1 Mi. 2MB is enough for about 12 seconds.
I think for jogging you need to go solid state. I think it's funny that people used to have armbands for their iPods back in the 2000s. iPod had mechanical hard drives in them. There were probably a bad choice until the iPod nano came out.
It's funny how memory used to be at a premium back then, and CD player makers were trying to use as little memory as possible and still get decent anti-skip. And now we have machines in our pockets that can hold an insane amount of albums on them. Did MiniDisc have issues with skipping?
From the manual of the D-E220 To maximize the skip protection performance (ESP MAX) The typical ESP (Electronic Shock Protection) function minimizes skipping by using a buffer memory that stores music data and plays it back in the event of a shock. The ESP MAX system is an extremely sophisticated technology. Due to the light weight and improved anti-shock characteristics of the optical block, this system recovers from the skipping much quicker than conventional technology such as ESP or ESP2.* * Although ESP MAX provides excellent protection, skipping may occasionally occur. It amazes me that CD manufacturers continued to iterate on CD anti-skip technology, rather than come up with a better format for portable listening that was solid state. Anyone remember SlotMusic from SandDisk? It was ennentially an album on an SD card in MP3 format. It was kind of a neat idea, but SD cards aren't really known for their longevity. It amazes me that I have CDs from 1982 that still play flawlessly. Sadly, I think the ship has sailed on a new physical media format. Everyone is into streaming these days.
Have a look at my post about the price of Compact Flash about halfway down this thread http://stereo2go.com/forums/threads/sony-network-walkman-personal-stereos.5800/ I wouldn't even consider myself an early adopter of Digital Cameras, getting my first in 2001 (a 3MPixel Canon that cost me £400). RAM wasn't cheap either. Someone in this thread http://www.tapeheads.net/showthread.php?t=23201 suggested 8MBytes would be sufficient, but I just knew that would be too expensive for a CD player. The expensive Toshiba Libretto computer running Windows only came with 8MBytes of RAM when it was released in 1996. Regarding jogging, anyone who claimed that a tape based player was best probably had a strong argument, although even the original iPod had 32MBytes so could buffer a decent amount of audio. To save power they don't even run the hard disk all the time, spinning it up when required and then shutting it down when the memory is full. I read that part of the spec is that all MiniDisc players had at least 10 seconds of buffer memory. Of course it would help that the data is already compressed on the disc, so that is one less thing to do, A memory the same size as the one in the Sony D-E226CK would give you about 15 seconds of buffering of the data straight off the MiniDisc.
RAM prices dropped almost to one tenth of the earlier pricing during 1996-97. The impossible became possible. My 1998 Sony Vaio has 256 MB of RAM.
At one point in the mid 1990s they reckoned RAM modules were worth more than their weight in gold. In the U.K. there was a problem with thieves stealing RAM. They would break into offices, rip open the computers and steal the RAM out of them. Far easier to carry than the actual computer. It happened in the offices opposite work
MiniDisc portables and Discmans are very similar, except for the write head present in the MiniDisc recorders. In some early MD units 4Mbit DRAM is present, later units have 16Mbit on board memory. MZ-R37 for instance was produced in 2 variants, one with 4 and one with 16Mbit for different markets. Actually the MZ-E55 and R55 use the same IC as the D-E400, a Panasonic MN41X17400CTT. Later units always buffer to memory, not just anti skip but primarily to keep power consumption during playback as low as possible, since the motors + laser are power hungry. Take for instance the MZ-N1: during cache read and playback 150mA, when cache is full and motors+laser off only 20mA. Looking at specified runtime on a single charge, you must conclude that most of the time the unit can run from cache, else it can never achieve more than 10 hours on a single gumstick (Max. 1350-1400mAh). After buffering motors and laser are turned off, until the cache is almost empty. Expressed in seconds for MD units, the cache can usually store 40s in standard play, 80s in LP2 and 160s in LP4.
According to Wikipedia "A buffer of at least six seconds is required on all MiniDisc players, be they portable or stationary full-sized units. This is needed to ensure uninterrupted playback in the presence of disc fragmentation." In contrast when CD was being designed 4MBits of memory would have been thought to be a huge amount. One of the selling points of the Commodore 64 was its "enormous" 64KByte (512KBit) memory. https://nosher.net/archives/computers/adve_072 An Anti-Skip CD player has to do a bit more than a MiniDisc as it has to compress the CD data in some way at the same time as playing it. Even a recordable MiniDisc only has to do one of the two tasks at a time.
Not sure if this is true for all MD players though. As example, a 80 minute disc formatted for HI-MD use can hold 300MB of music, compressed ATRAC or linear PCM. The 40 seconds buffer needs to fit in 2MB (16Mbit per service manuals). That means 50KB / seconds at most, which equals maximum 400Kbit/s. Likely that the source was not always of that high quality, in which case the cache could hold more than 40 seconds. 400Kbits is actually not that bad, compared to other formats like MP3. However, at 50KB / second a 80 minute MiniDisc would contain 240MB of data, while it can hold 300MB? Perhaps formatted as normal MD it can only hold 240MB , in which case no encoding is needed prior to storing in cache. 1:1 relation to data read and fed into cache, decoding after reading to cache. Note that this cache size is pretty common, even higher end Sony CD portables use it (D-EJ825 -925/ D-EJ888 / D-EJ925). I could not find the seconds of anti skip cache as marketed, but at 40 seconds the source needs to be highly compressed. An audio CD of 80 minutes is 700MB, which means that 2MB as Discman cache can never contain more than 13 seconds uncompressed. The 6 seconds is the true buffer for continuous playback even on stationary Minidisc machines, due to storage technique used. It is not related to anti skip as used in Discmans, or later MD units where it is actually referred to a G-Protection (see MZ-N710 for instance).
The first Anti-Skip CD players only stored 3 seconds of music in memory anyhow. I just found a good article from 2000 back when this was the latest thing https://www.nytimes.com/2000/03/16/...at-last-cd-players-that-d-d-d-don-t-skip.html the information about fast focussing etc is interesting. I have found that portable CD players like the D-E220 are far less prone to skipping even with ESP turned off. The title of the article brings back memories of being in pubs when the CD player would reach a dirty part of the disc and S-S-S-S-S-S-Stick, sometimes requiring a thump on the player by a member of bar-Staff to free it.
Thanks @Longman for the additional background information. I meanwhile found evidence of a certain type of compression with G-Protection set to 2 on a D-EJ925. Per the operating instructions of this series of players, you can enjoy high quality music while walking with the switch set to 1. However, while jogging you need to set to switch to 2 for better protection against skipping. Note the absence in that line of ‘high quality’, from which I derive that there is compression applied to ensure the cache can hold a longer part of the song at the cost of quality. Side note: the switch is inconveniently located underneath the CD, while there are also other switched at the bottom (outside) of the player. Sony could have easily squeezed the 3 switch next to the 2 already present.
I recently discovered the PMD-650 by Marantz (Philips), at least online. It seems that this professional recorder features 16MB (not Mb) for 20 seconds of stereo buffer, or 40 seconds mono.
