First, I would like to say "Thank you" for doing this experiment and for reporting the results, including updates! You have chosen to use an enterprise-class SSD in your TiVo and it appears that you are using about 1% of the unit's endurance each month. Endurance ratings for enterprise-class SSDs are different than those for consumer-grade products, and, in fact, they seem to more closely match the operating environment seen inside the Bolt's case. Specifically, the JEDEC SSD standards organization defines the enterprise operating temperature to be 55°C versus 40°C for consumer grade SSDs, as can be seen on page 26 of this presentation by the then-acting chairman of the SSD standards committee. In addition, the enterprise-grade SSDs ratings are made by operating the units 24 hours/day versus 8 hours/day for consumer SSDs. Again, this seems like a good match for the TiVo Bolt. Also, these enterprise-class SSDs have much higher endurance ratings than their consumer-grade counterparts. This all seems very good for the idea of using these enterprise-class SSDs in a TiVo, so where is the rub? If there is one (and I'm not sure there is), I would have to say it has to do with how JEDEC defines the end of life of these SSDs. In fact, the failure criteria for enterprise and consumer SSDs are different, as can also be seen on page 26 of that JEDEC presentation. Notably, the end-of-life criterion for data retention for consumer-grade SSDs is one year, but for enterprise-level SSDs it is only three months! On page 27 of the presentation they show two tables which include the MODELED number of WEEKS of data retention remaining at the end of life of SSDs. The top table shows remaining data retention for a consumer-grade SSD at the end of its life and the bottom table shows it for an enterprise-grade SSD. The green boxes show the measurement criteria for those two classes of SSDs. So, what to make of all of this? Here are some of my conclusions based on the data in this presentation: - Most importantly, one of the ways that SSDs degrade is in their ability to retain the data which has been written to them. - I don't see any information about how long SSDs retain data when new, but it seems that will depend upon the temperature at which the data was written. - One of the reasons enterprise-class SSDs have higher endurance ratings is that their end-of-life criterion for data retention is more lax than is the one for consumer-grade SSDs. You can see this by looking at the two tables on page 27 of the presentation: The end-of-life consumer SSD still has 52 weeks of data retention by the enterprise criterion while the end-of-life enterprise SSD only has 10 weeks (2.5 months!) of data retention by the consumer criterion. - It seems clear that a KUID recording on an SSD cannot simply be written and left there for the years that the user might expect it to be available. For instance, if we assume a 55-degree operating temperature for the end-of-life enterprise-class SSD, you can see that anything written on such a degraded unit must be reread and rewritten within two weeks or the information will be lost (top right cell in enterprise chart). - For a 24-hour-per-day TiVo application (I understand this is NOT what the OP is doing), the temperature of the SSD will be fairly constant. As such, it seems that for the purpose of data retention while operating, cooler temperatures are preferred since storage and write temperatures are the same in that case (see the bottom-left to top-right diagonal cells). All this brings up the following questions in my mind: - Does the SSD firmware interpolate into a set of these degradation tables while operating to determine how often to reread and rewrite data before it gets lost, based on total data written and temperature? I assume that it would have to. If it only used the temperature information for that purpose, then it would need to rewrite data too frequently, IMO. - Does operation of the SSD at high temperature cause it to have more write endurance? According to this source, it does: The obvious implication is that it would be best to keep your SSD buttoned up in your Bolt and running hot to improve its endurance. Too bad everything else in the box will suffer! - I wonder at what percentage of life the OP's SSD will have a powered-off shelf life of only one year like a consumer-grade unit would have at end of life. - Would a consumer-grade SSD have the same (or better?) life as an enterprise-grade SSD in a TiVo application? I ask this because the Micron 5210 series SSD the OP chose uses 4-bits-per-cell technology (it must write and read 16 different charge levels in each cell). This is equivalent to the Samsung QVO series, which is their lowest-grade consumer product. The Samsung 860 EVO series uses 3 bits per cell, meaning that it only has to differentiate between 8 voltage levels while the Samsung 860 PRO uses only 2 bits per cell, or 4 different voltage levels. I would think the Samsung 860 Pro would have both much more endurance AND much better data retention. But I wonder if TiVo has already rendered this entire discussion moot by preventing SSDs from being used in Bolts running TE4. Does anyone have an SSD running in a TE4 Bolt? If so, please let us know your experience. Thoughts?