For our power consumption testing, we have the drive connected to the system as a secondary drive. To record the wattage, we use an Amprobe AM-270 multimeter connected in line with the 5v power on our SATA power cable to the drive. The multimeter records the min/max amperage draw from the drive over our testing period.
We also record the drive’s sequential and random read and write power draw using Iometer. We then take the values recorded and calculate the wattage of the drive. Some of the results may seem high compared to a standard notebook HDD because as these are peak values under load. When we see average power draw, SSDs are still more power efficient because they only hit max power for a short period of time.
The 480GB Patriot Ignite’s power consumption is rated for 2.61W read, 5.54W write, and 0.6W idle. In our testing we found that while at idle the Ignite did much better than its rating and only consumed 0.275W. Sequential read followed and consumed less than the rated spec at 2.31W, however, once we ran sequential writes the drive started to consume 5.690W. This is only 150mW over rated specs which is okay, but then we saw a max draw of 6.02W. With this and lack of DevSleep support, the Ignite is definitely not a great option for those with Ultrabooks or mobile platforms where every bit of juice counts.
REPORT ANALYSIS AND FINAL THOUGHTS
Thinking back to our briefing with Patriot at CES, I have to admit that getting that first look Patriot Ignite was refreshing as it’s the first new high-end SSD that Patriot has released in a while. We definitely had high hopes as the Phison S10 controller brings a lot to the table. Based on our results, we are definitely getting a new look at asynchronous SSD performance, just not enough unfortunately to lift the Patriot Ignite from the mid-tiered SSD category. Lack of DEVSLP and the 3 year warranty is also what we typically see with this caliber of SSD.
As much as we have definitely seen an amazing improvement in asynch NAND performance with Micron’s newest 16nm design, the difference in NAND types became evident in PCMark 8 testing. While high sequential speeds in benchmarks paint a picture of great performance, lower 4K speeds with incompressible data and underwhelming PCMark results paint another for real world usage.
On the flip side once again, we are able to see just how far performance has increased from previous generations of asynchronous NAND SSDs. Never would we have thought this drive would be able to achieve the read and write numbers it is rated for with incompressible data, nor would we have expected to see 100K IOPS read performance out of it! Not many SSDs can hit that 100K mark, let alone with asynchronous NAND!
While the Ignite may not perform as well as those few big dogs, it is still a fairly solid offering, especially when you consider that Patriot has been out of the SSD game for a while. Alas, the Ignite also contains many error correcting features along with end-to-end data path protection to aid in data reliability, and it does have AES 256-bit encryption to top that off, unlike many others.
Based upon our impressions and testing results from the Patriot Ignite, we award it our Bronze Seal!
Check Out the Patriot Ignite at Amazon Today!
Good article, but I just noticed what I think is a typo from your spell checker.
“Alas, the Ignite also contains many error correcting features along with
end-to-end data path protection to aid in data reliability, and it does
have AES 256-bit encryption to top that off, unlike many others.”
From the context, I’d guess you meant “Also” instead of “Alas”. “Alas” implies regret, disappointment, sadness, and so on.
I wouldn’t agree with the reviewer’s statements about async NAND causing the lower bandwidth in the PCMark recovery phases, as its latency didn’t rebound properly either.
Async NAND normally has reduced peak read bandwidth vs sync, but in this case the Phison controller’s parallel channel operations or some form of interleaving along with those features avoid this as seen in the AS-SSD tests which can’t be fooled by the controller’s data compression acceleration.
The reduced bandwidth and increased latency in the recovery stages is most likely either a marginal TRIM implementation or a very lazy one to avoid big stutters / latency from the drive doing garbage collection, which instead appears to cause longer-term performance degradation. Since we never seen those numbers properly settle like the other drives, how long the TRIM process takes to fully complete is unknown. TRIM is drive-controlled vs OS controlled–the OS simply passes the command (if supported) and then the drive is on its own to do the TRIM processing at its leisure. In this case it seems like the drive’s firmware performs TRIM very slowly as a background process, which affects both drive response and peak bandwidth (despite it being a quad core controller).
Unfortunately Patriot is very bad for releasing SSD firmware updates in a timely matter (if at all). Despite the impressive ability of the Phison S10 controller to minimize the compromises of async vs sync flash, I’d buy this architecture from another manufacturer if another product comes along at a similar price until Patriot proves itself to be taking the SSD market seriously again. A classic example of Patriot’s behavior is that 3+ years later, the finally released firmware beyond the TRIM-broken 5.02 firmware (TRIM-fixed 5.04 was released and then pulled) for their Pyro Sandforce based drives a few months back. This is YEARS too late, and their site only provides a download for their Pyro drives; their other (higher end) Sandforce-based drives which should use the identical firmware release (Pyro SE and Wildfire) are no longer even listed in the downloads section. Apparently even spending a premium for Patriot’s higher end offerings means nothing to them, so I’ve basically blacklisted using them for any of my work until they prove themselves worthy of anything more than a plug-n-run-away install.
Silicon Power produces drives with Phison controllers (albeit randomly interchanged with SF-based drives under the same product name), and they have a much better track record for supporting them with firmware updates. If they produced a similar S10-based drive, that would be my recommendation.