Sorry to be straight forward but is Paypal the only reason why you announced this 3-4 weeks delay from your latest inputs? And where are the risks for this schedule?
Please do not misunderstand, this is not to complain about the new target (although I sure had wish to get my V in April the latest) and I am not denying either the seriousness of the PP issue. However somehow it doesn’t fit the information you gave us before. I also want to understand how safe is the new target.
From all the piece of information you shared before, I understood that :
@Konstantinos was saying in Early February ;
In the video updates from feb 16th (from 3’) @Konstantinos mentions end of March/1st week of April, this being the vendors information (assembly line)
In the same video @Konstantinos also says “Now we already have all the components purchased”.
Now you are coming back saying “we have a paypal issue and we could thus not ordered some long leads components such as the intel ssd.”
For me this is not consistent with your previous statements. Somehow all long TAT components should have been purchased first.
Since you mentioned also that you did not informed us immediately, when did the PP issue started? Since all final payments occurred on Feb 28th, May I expect that this is the starting date? The mail that you give as example is from March 7th. Is the delay due to this late change to intel SSD?
I don’t know where it was written, but @Konstantinos mentioned an issue with PayPal some time ago, but said that they will inform us about it, if needed. This fits to him saying now that the issue started quite some time ago and they wanted to handle it privately with PayPal.
From what I learned so far, it seems that DRAM-less cache-less SSD like the SanDisk Z400s would give the best battery life, although you seem to need some solution to prevent write amplification that comes from the lack of DRAM. This is my theoretical solution to solve the issue.
EDIT It seems that Windows has RAM caching built in, anybody can confirm? If thats the case, then why do SSDs come with DRAM?
If outright performance is the top priority, the crown easily goes to the Samsung PM961 (similar to 960 EVO), hands down.
If the PM961 is too expensive and the DRAM replacement is too difficult to implement, then the good ol’ 850 EVO seem to be the jack of all trades. In addition to that, it seems to be less power hungry than the PM961 or any other PCIe SSD for that matter.
However, if PCIe is a must, here’s a list of some of the most popular PCIe SSDs, including the 600p and the aforementioned PM961. The graph shows the battery life on the Lenovo Y700 when equipped with each SSD. In term of performance, almost every single one of them is faster than the Intel 600p, thats more than sufficient for the Eve V in my opinion.
Both the Plextor and WD have Marvell 88SS1093 controller, leading me to believe that this controller is the key to a power efficient SSD.
It is probably also worth taking a look at the ADATA XPG SX8000, although I dont have any information on its power consumption. It might be similar to the 600p since they share the same controller, the main difference might come from the fact that the ADATA doesnt have separate, huge SLC cache. ADATA instead implemented SLC-mode in the main MLC NAND, buffered by a DRAM that the 600p lacks. For the results, the (claimed) read and write speed both favors the SX8000 over the 600p.
tldr Best battery life (SATA): SanDisk Z400s (or other DRAM-less cache-less SSD) Best battery life (PCIe): tie between Plextor M8Pe and WD Black Best performance (SATA): Samsung 850 EVO Best performance (PCIe): Samsung PM961
Performance: Z400s < 850 EVO < 600p < WD Black < Plextor M8Pe < PM961 Efficiency: 600p < PM961 < WD Black = Plextor M8Pe < 850 EVO < Z400s (Intel is kind enough to provide engineering support, this might change the position of 600p here)
Personally I would go with the SanDisk if the DRAM issue can be resolved, since I stand by the original consensus that faster SSD doesnt give a significant performance benefit on this type of devices, while 1-2 hours of battery life does.
EDIT I was comparing the 240-256 GB models of each SSD since it is the middle ground of the V. Upon further research, the SanDisk is not available in 500 GB variant, and the PCIe options I presented above (ADATA SX8000, Plextor M8Pe, WD Black, Samsung PM961) are quite a lot more expensive than the Intel 600p in the retail market, leaving 600p as a good performance-per-dollar value among 500 GB PCIe SSDs. However, we should not overlook the above-average power consumption that seems to plague the 600p. Unless Intel’s optimizatio could mitigate this problem, I would still recommend the 850 EVO over the 600p if it could bring an additional 30-60 minutes of battery life. Also dont forget that Eve might get special bulk discount from either of those companies, so far we only know that Intel cannot provide one. If nobody gives special bulk discount, we technically could even have a poll to have different OEM for each m3, i5, and i7 model. For example, m3 buyers might prioritize battery life over performance, and the other way around for i7 buyers. This makes the Z400s a possibility for the m3 and i5 models.
@Patrick_Hermawan: Thanks a lot for this extensive review.
As a layman in this respect I would only like to add that I do not see these large differences in the battery life graphs - most of them are insignificant (950 Pro as an exception).
In graph 1 between 600p and the SM961 there is a insignificant difference - only if you go to the WD Black or the M8Pe the difference is somehow noticeable.
In graph 2 the group of disks between the 600p and the SM961 is basically on the same level.
So in this sense I do not see a need for change - the other factors such as performance and the other technical factors aside.
The difference is insignificant in the test system, which is a Lenovo Y700, because the rest of the system uses a lot more power than it is on the Eve V. As an example, the screen has almost double the total area, the CPU is 10x the TDP, and so on. Therefore, the SSD power consumption plays a significantly smaller role than it is on the V.
In addition to that, since the Y700 has short battery life to begin with, a small percentage of difference on this system is easily dismissible. 8% of battery life on the Y700 is just a mere 15 minutes, which is easily negligible to most people. Meanwhile, 8% of battery life on the V is already an hour! As demonstrated before, even a fraction of Watts can mean an hour to the V. See my first calculation here Latest updates from China on prototypes, deliveries and more!
This sort of problem is what you would face when youre building a highly mobile portable device like phones, tablets, Ultrabooks, MP3 players, eBook readers, smartwatches, etc. Every miliwatts count here, and it would be a shame if all Eve Team’s hard work to optimize the battery life is wasted just by a power-hungry SSD alone.
And finally, as you mentioned, while some SSDs I mentioned consume a similar amount of power as the 600p, we shouldnt forget the fact that most of them can easily smoke the 600p in term of performance.
Can I just ask that those who prefer performance not be forgotten about? Battery life can be replenished by a charge. But once you sacrifice that performance, that’s it you can’t do anything to get it back.
I’d like to see the V ship with a pcie if possible and whilst I fully understand the benchmarks etc, I’d like to see some benchmarks from on the V. As having direct access to engineering support and optimisation can make a huge difference.
There was an update about 2 days ago. Since then there was a Sunday and today. People are working on it, but it was mentioned that alternative financing was made available to not do further harm to the timeline.
This means: no further update as of now, but we’ll hear as soon as there’s something to be told.
And on a personal note: asking a question in all caps is rude! Please don’t do that. THANK YOU!
The funds are (most probably) still frozen. There hasn’t been an update on that. Again, it was weekend and I would never expect anything to get resolved on a weekend when it involves communication between different companies (and particularly banks on top).
What do you mean by “I’m not in”? There’s currently nothing the community has to worry about explicitely - unless asked for by Eve. There is communication between Eve and Paypal and I am in good faith that they’ll sort things out.
Again, Eve takes care of the funds. They also took care of getting alternative funds to allow all payments to take place. From a user/backer perspective there’s nothing to be worried about.
Potential loan fees have not been considered before, but this is something I see paypal liable for in the long run (and I assume that a legal team will work that topic out).
Once the funds are released they will offset the bank loan. Thus, that is nothing that will affect the users/backers.
Eve has the feedback from IGG and backerkit on who paid what amount. I as a backer cannot be made liable for funds frozen on the other end. I can prove that I have done my duty and expect my device to be shipped once the launch is confirmed.
To be quite frank, it is not my business whether Eve has trouble in getting the funds released. (That does not mean that I don’t care as I want Eve to be successful.)
Of course to each their own, but you have to have certain trust as well. As you try to get a swap spot your funds would anyways go to someone who has paid Eve, so your money does not go into the frozen funds.
Part goes to the buyer, the rest goes to EVE. I half-way completed such a transaction. Regardless of the amount paid by the buyer, the balance was due to EVE by 2-28 (maybe that’s been modified; I don’t know).
I will watch your progress - I wish you luck! Such a financial issue is a troubling “red flag” to me. If they are successful in releasing units, I will again consider purchasing, but not until then.