If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Rate Thread | Display Modes |
#1
|
|||
|
|||
Samsung Magician
There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene |
Ads |
#2
|
|||
|
|||
Samsung Magician
On 6/29/19 4:16 PM, Rene Lamontagne wrote:
There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene Hi Rene, I don't actually know, but you can ask them directly at 800-726-7864. Their tech support is pretty good. -T |
#3
|
|||
|
|||
Samsung Magician
On 6/29/19 6:29 PM, T wrote:
On 6/29/19 4:16 PM, Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene Hi Rene, I don't actually know, but you can ask them directly at 800-726-7864. Their tech support is pretty good. -T Or eMail them at |
#4
|
|||
|
|||
Samsung Magician
On 2019-06-29 8:30 p.m., T wrote:
On 6/29/19 6:29 PM, T wrote: On 6/29/19 4:16 PM, Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene Hi Rene, I don't actually know, but you can ask them directly at 800-726-7864. Their tech support is pretty good. -T Or eMail them atÂ*Â* Will try next week. Thanks, Rene |
#5
|
|||
|
|||
Samsung Magician
Rene Lamontagne wrote:
There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? Googleing returns me some good results and a lot of Not sure results. I believe it uses cache memory, for what it's worth. If the reader wants fast Windows and data storage, buy one or two NVMe (pronounced "envy me") drives. Use your other drives for backup. |
#6
|
|||
|
|||
Samsung Magician
Rene Lamontagne wrote:
There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene How will you know it's bulletproof ? Intel tried to build one of these years ago, and after a couple years of bug busting, they gave up on theirs. Paul |
#7
|
|||
|
|||
Samsung Magician
On 2019-06-29 8:37 p.m., John Doe wrote:
Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? Googleing returns me some good results and a lot of Not sure results. I believe it uses cache memory, for what it's worth. If the reader wants fast Windows and data storage, buy one or two NVMe (pronounced "envy me") drives. Use your other drives for backup. Yes I have Adata SX8200 NvMe,s these are the leftovers from my new build and are repurpused now, they are plenty fast, I was just curious about the Rapide mode thingy. Rene |
#8
|
|||
|
|||
Samsung Magician
On 2019-06-29 8:56 p.m., Paul wrote:
Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene How will you know it's bulletproof ? Intel tried to build one of these years ago, and after a couple years of bug busting, they gave up on theirs. Â*Â* Paul After reading a few articles on how it works and their use of your memory I decided that it really wasn't worth playing with and now have lost interest. Rene |
#9
|
|||
|
|||
Samsung Magician
It`s a RAM cache. If you would like to play with a RAM cache, use the
best available (4 weeks free or so): https://www.romexsoftware.com/en-us/primo-cache/ Most interesting is the write cache, I use 30 seconds for buffering before writing to SSD. 4 GB total for Primocache, 1 GB for reading, 3 GB for writing. The system has about 4-5 GB more free for caching by Windows. Regards M. On Sat, 29 Jun 2019 18:16:45 -0500, Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene |
#10
|
|||
|
|||
Samsung Magician
Rene Lamontagne wrote:
There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene I've used it since my Win7 setup when I got the Samsung EVO-840. Their benchmarks are skewed. Other benchmarks weren't so impressive, but they were better with Rapid Mode on. However, you need to running into those edge-use cases where a RAM cache will help. I don't do huge video editing to make use of such huge write buffers. Rapid Mode is on my To Do list to disable. I don't remember if I enabled it in my new Win10 build. If I did, I'll disable it when it get to that item way down in my computer To Do list. As far as I can tell, it hasn't hurt me; however, it hasn't helped, either, so there's no point in keep using it. http://www.thessdreview.com/software...e-2-1-testing/ (dated December 19, 2014) Samsung˙s RAPID white paper states that RAPID works by analyzing ´system traffic and leverages spare system resources (DRAM and CPU) to deliver read acceleration through intelligent caching of hot data and write optimization through tight coordination with the SSD.ˇ It's one of these whiz-bang features that sound great on paper and get promoted, like how an $800 water-filtering vacuum is so much cleaner than a $190 dry vacuum, providing you can past the junky watery mess you have to flush down the toilet while hoping you don't plug it. When I watch those commercials, first thing that comes to my mind is "Why aren't the dry vacuum users cleaning more often, so they don't have to pay $800 to get their carpets clean?" I have a 3-second delay to show the multiple boot choice on startup. I use Macrium Reflect and have it added as a boot-time option, so I need to restore than I don't have to go hunting for a bootable CD or USB drive to bring up Reflect. I won't notice another half-second for boot time. However, Samsung's claim is that boot time would be reduced, not increased; else, they'd lose a selling point right there. For the application loat-time test, I opened the Before and After images from the article, and toggled between those two tabs in the web browser. I did see some difference, but it was miniscule. Excel was the only one that benefited from Rapid Mode. Photoshop was the one with the biggest change (and biggest was still a tiny change); however, the first file open was faster with Rapid Mode but subsequent file opens were slower. No win there for Rapid Mode. Besides, who keeps loading and unloading data files as fast as they can? There is no read performance boost from Rapid Mode yet it does add another layer of vulnerability (to failure, not from infection). This is like adding sTunnel to an old version of Outlook that doesn't have TLS support required by the chosen e-mail provider, just making the chain longer where each added link increase the failure potential. A couple tests had Rapid Mode the winner; however, I'm not running my home PC as a file or application server. Personally, I prefer a more stable platform. That's why I don't overclock, too. Availability and reliablity are more important to me than a few edge-cases I may never encounter to effect a performance boost as a rare-time event. Remember those hybrid HDDs that used SSDs for caching? Yeah, they were faster than HDD-only drives but cost a lot more and I didn't hear any of its users extoling the benefits after a year's real-world experience with them. System RAM is faster than flash memory, so the same concept was applied to SSDs: speed them up with a RAM cache. I kept using HDDs until I could afford to jump to SSDs rather than spend money on incremental hardware changes. As for RAM caching of SSDs, naw, I'll wait until something comes along that gives real-world and consistent performance boost in some new hardware advance. Memristor's anyone? If they ever do show up, it'll be another 10 years before they hit a price point tolerable to consumers. I'll be dead by then. I'm not running a transaction server used by hundreds of customers where even the tiniest fraction of performance increase increases revenue that will reduce ROI and start generating revenue. I haven't any wow-factor performance boost in my Win7 (8GB) and Win10 (64GB) with Rapid Mode enabled. No point in throwing more software and hardware at a solution that doesn't effect any real or experienced benfit. Also, whether it is cost effective depends on how much system memory you have. One a 4GB system, nope, won't help and probably won't work. On 8GB, I still wouldn't waste the RAM. With 32 or 64GB then maybe I'd be willing to waste some RAM on caching the SSD, but by then I'd probably be using a RAMdisk to make a super-fast drive that is far faster than is flash memory and doesn't self-destruct over time due to oxide stress or write volume concern. The only reason why I will keep Samsung Magician is to change the overprovisioning of the SSD; however, apparently just enlarging the unallocated space on the drive will do that, too. If I find that last claim is true, bye bye Samsung Magician (after first using it to disable Rapid Mode). |
#11
|
|||
|
|||
Samsung Magician
On 2019-06-30 5:06 a.m., VanguardLH wrote:
Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene I've used it since my Win7 setup when I got the Samsung EVO-840. Their benchmarks are skewed. Other benchmarks weren't so impressive, but they were better with Rapid Mode on. However, you need to running into those edge-use cases where a RAM cache will help. I don't do huge video editing to make use of such huge write buffers. Rapid Mode is on my To Do list to disable. I don't remember if I enabled it in my new Win10 build. If I did, I'll disable it when it get to that item way down in my computer To Do list. As far as I can tell, it hasn't hurt me; however, it hasn't helped, either, so there's no point in keep using it. http://www.thessdreview.com/software...e-2-1-testing/ (dated December 19, 2014) Samsung˙s RAPID white paper states that RAPID works by analyzing ´system traffic and leverages spare system resources (DRAM and CPU) to deliver read acceleration through intelligent caching of hot data and write optimization through tight coordination with the SSD.ˇ It's one of these whiz-bang features that sound great on paper and get promoted, like how an $800 water-filtering vacuum is so much cleaner than a $190 dry vacuum, providing you can past the junky watery mess you have to flush down the toilet while hoping you don't plug it. When I watch those commercials, first thing that comes to my mind is "Why aren't the dry vacuum users cleaning more often, so they don't have to pay $800 to get their carpets clean?" I have a 3-second delay to show the multiple boot choice on startup. I use Macrium Reflect and have it added as a boot-time option, so I need to restore than I don't have to go hunting for a bootable CD or USB drive to bring up Reflect. I won't notice another half-second for boot time. However, Samsung's claim is that boot time would be reduced, not increased; else, they'd lose a selling point right there. For the application loat-time test, I opened the Before and After images from the article, and toggled between those two tabs in the web browser. I did see some difference, but it was miniscule. Excel was the only one that benefited from Rapid Mode. Photoshop was the one with the biggest change (and biggest was still a tiny change); however, the first file open was faster with Rapid Mode but subsequent file opens were slower. No win there for Rapid Mode. Besides, who keeps loading and unloading data files as fast as they can? There is no read performance boost from Rapid Mode yet it does add another layer of vulnerability (to failure, not from infection). This is like adding sTunnel to an old version of Outlook that doesn't have TLS support required by the chosen e-mail provider, just making the chain longer where each added link increase the failure potential. A couple tests had Rapid Mode the winner; however, I'm not running my home PC as a file or application server. Personally, I prefer a more stable platform. That's why I don't overclock, too. Availability and reliablity are more important to me than a few edge-cases I may never encounter to effect a performance boost as a rare-time event. Remember those hybrid HDDs that used SSDs for caching? Yeah, they were faster than HDD-only drives but cost a lot more and I didn't hear any of its users extoling the benefits after a year's real-world experience with them. System RAM is faster than flash memory, so the same concept was applied to SSDs: speed them up with a RAM cache. I kept using HDDs until I could afford to jump to SSDs rather than spend money on incremental hardware changes. As for RAM caching of SSDs, naw, I'll wait until something comes along that gives real-world and consistent performance boost in some new hardware advance. Memristor's anyone? If they ever do show up, it'll be another 10 years before they hit a price point tolerable to consumers. I'll be dead by then. I'm not running a transaction server used by hundreds of customers where even the tiniest fraction of performance increase increases revenue that will reduce ROI and start generating revenue. I haven't any wow-factor performance boost in my Win7 (8GB) and Win10 (64GB) with Rapid Mode enabled. No point in throwing more software and hardware at a solution that doesn't effect any real or experienced benfit. Also, whether it is cost effective depends on how much system memory you have. One a 4GB system, nope, won't help and probably won't work. On 8GB, I still wouldn't waste the RAM. With 32 or 64GB then maybe I'd be willing to waste some RAM on caching the SSD, but by then I'd probably be using a RAMdisk to make a super-fast drive that is far faster than is flash memory and doesn't self-destruct over time due to oxide stress or write volume concern. The only reason why I will keep Samsung Magician is to change the overprovisioning of the SSD; however, apparently just enlarging the unallocated space on the drive will do that, too. If I find that last claim is true, bye bye Samsung Magician (after first using it to disable Rapid Mode). After reading your post VanguardLH I decided to take One more kick at the cat. I also have Macrium Reflect on my start multi-boot menu, Makes life easier, I decided on a real world write test which I do quite often which is Do a Macrium backup of my C:\ drive to the Samsung 850 evo. these are the results with Rapide mode *OFF* Size 31.6 GB, time 2:20 minutes, CPU use 19%, memory use 2.4 GB. These are the results with Rapide Mode *ON* Size 31.6 GB, time 2.05 minutes, CPU use 21%, memory use 6.5 MB. If my math is still functional that's about a 7 % gain which is about 4 seconds, not worth bothering with, So In my case Rapide mode will remain off permanently. Rene |
#12
|
|||
|
|||
Samsung Magician
On 2019-06-30 10:08 a.m., Rene Lamontagne wrote:
On 2019-06-30 5:06 a.m., VanguardLH wrote: Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene I've used it since my Win7 setup when I got the Samsung EVO-840.* Their benchmarks are skewed.* Other benchmarks weren't so impressive, but they were better with Rapid Mode on.* However, you need to running into those edge-use cases where a RAM cache will help.* I don't do huge video editing to make use of such huge write buffers. Rapid Mode is on my To Do list to disable.* I don't remember if I enabled it in my new Win10 build.* If I did, I'll disable it when it get to that item way down in my computer To Do list.* As far as I can tell, it hasn't hurt me; however, it hasn't helped, either, so there's no point in keep using it. http://www.thessdreview.com/software...e-2-1-testing/ (dated December 19, 2014) ** Samsung˙s RAPID white paper states that RAPID works by analyzing ** ´system traffic and leverages spare system resources (DRAM and CPU) to ** deliver read acceleration through intelligent caching of hot data and ** write optimization through tight coordination with the SSD.ˇ It's one of these whiz-bang features that sound great on paper and get promoted, like how an $800 water-filtering vacuum is so much cleaner than a $190 dry vacuum, providing you can past the junky watery mess you have to flush down the toilet while hoping you don't plug it.* When I watch those commercials, first thing that comes to my mind is "Why aren't the dry vacuum users cleaning more often, so they don't have to pay $800 to get their carpets clean?" I have a 3-second delay to show the multiple boot choice on startup.* I use Macrium Reflect and have it added as a boot-time option, so I need to restore than I don't have to go hunting for a bootable CD or USB drive to bring up Reflect.* I won't notice another half-second for boot time.* However, Samsung's claim is that boot time would be reduced, not increased; else, they'd lose a selling point right there. For the application loat-time test, I opened the Before and After images from the article, and toggled between those two tabs in the web browser. I did see some difference, but it was miniscule.* Excel was the only one that benefited from Rapid Mode.* Photoshop was the one with the biggest change (and biggest was still a tiny change); however, the first file open was faster with Rapid Mode but subsequent file opens were slower. No win there for Rapid Mode.* Besides, who keeps loading and unloading data files as fast as they can? There is no read performance boost from Rapid Mode yet it does add another layer of vulnerability (to failure, not from infection).* This is like adding sTunnel to an old version of Outlook that doesn't have TLS support required by the chosen e-mail provider, just making the chain longer where each added link increase the failure potential. A couple tests had Rapid Mode the winner; however, I'm not running my home PC as a file or application server.* Personally, I prefer a more stable platform.* That's why I don't overclock, too.* Availability and reliablity are more important to me than a few edge-cases I may never encounter to effect a performance boost as a rare-time event. Remember those hybrid HDDs that used SSDs for caching?* Yeah, they were faster than HDD-only drives but cost a lot more and I didn't hear any of its users extoling the benefits after a year's real-world experience with them.* System RAM is faster than flash memory, so the same concept was applied to SSDs: speed them up with a RAM cache.* I kept using HDDs until I could afford to jump to SSDs rather than spend money on incremental hardware changes.* As for RAM caching of SSDs, naw, I'll wait until something comes along that gives real-world and consistent performance boost in some new hardware advance.* Memristor's anyone?* If they ever do show up, it'll be another 10 years before they hit a price point tolerable to consumers.* I'll be dead by then. I'm not running a transaction server used by hundreds of customers where even the tiniest fraction of performance increase increases revenue that will reduce ROI and start generating revenue.* I haven't any wow-factor performance boost in my Win7 (8GB) and Win10 (64GB) with Rapid Mode enabled.* No point in throwing more software and hardware at a solution that doesn't effect any real or experienced benfit. Also, whether it is cost effective depends on how much system memory you have.* One a 4GB system, nope, won't help and probably won't work.* On 8GB, I still wouldn't waste the RAM.* With 32 or 64GB then maybe I'd be willing to waste some RAM on caching the SSD, but by then I'd probably be using a RAMdisk to make a super-fast drive that is far faster than is flash memory and doesn't self-destruct over time due to oxide stress or write volume concern. The only reason why I will keep Samsung Magician is to change the overprovisioning of the SSD; however, apparently just enlarging the unallocated space on the drive will do that, too.* If I find that last claim is true, bye bye Samsung Magician (after first using it to disable Rapid Mode). After reading your post VanguardLH I decided to take One more kick at the cat. I also have Macrium Reflect on my start multi-boot menu, Makes life easier, I decided on a real world write test which I do quite often which is Do a Macrium backup of my C:\ drive to the Samsung 850 evo. these are the results with Rapide mode *OFF* Size 31.6 GB, time 2:20 minutes, CPU use 19%, memory use 2.4 GB. These are the results with Rapide Mode *ON* Size 31.6 GB, time 2.05 minutes, CPU use 21%, memory use 6.5 MB. If my math is still functional that's about a 7 % gain which is about 4 seconds, not worth bothering with, So In my case Rapide mode will remain off permanently. Rene In my mind it was about 4 seconds, using a calculator it is actually 9.8 seconds, My old age is showing. :-) Rene |
#13
|
|||
|
|||
Samsung Magician
Rene Lamontagne wrote:
On 2019-06-30 10:08 a.m., Rene Lamontagne wrote: After reading your post VanguardLH I decided to take One more kick at the cat. I also have Macrium Reflect on my start multi-boot menu, Makes life easier, I decided on a real world write test which I do quite often which is Do a Macrium backup of my C:\ drive to the Samsung 850 evo. these are the results with Rapide mode *OFF* Size 31.6 GB, time 2:20 minutes, CPU use 19%, memory use 2.4 GB. These are the results with Rapide Mode *ON* Size 31.6 GB, time 2.05 minutes, CPU use 21%, memory use 6.5 MB. If my math is still functional that's about a 7 % gain which is about 4 seconds, not worth bothering with, So In my case Rapide mode will remain off permanently. Rene In my mind it was about 4 seconds, using a calculator it is actually 9.8 seconds, My old age is showing. :-) Rene The reason this happens, is Macrium is rate-limited by its need to calculate a checksum as it writes out the backup. I believe Macrium could go a lot faster, if there was no requirement for a checksum. But the checksum (hash) that it does, is important, because that's what makes a Verify run work. You can Verify a .mrimg using Macrium and determine whether any byte corruptions have happened. For example, if you have five .mrimg backups, you test all of them and one is corrupt, you can toss it in the trash, and that will leave room for another fresh backup some day. I've had two Macrium backups fail to verify, and the root cause was bad RAM (since replaced). And that's why we don't ask Macrium to remove the hash step. Now, with your whiz-bang 8700, your backup probably goes twice as fast as mine. I can probably manage 285MB/sec when the storage subsystem limitations have been removed. If you tried some other fake test cases, you could probably make an argument that "it's a big win" or something :-) If the Samsung Cache could use LargePages, then the scheme *would* be a big win. For example, some processors support a couple LargePage entries of 1GB mapping. So a table-walk only happens once for each gigabyte of caching being done. Normal page table size is 4KB, and the page table is constantly doing table-walks as the cache would fill. This is what helps "prevent RAM from being fast", is the tiny size of the page table entries. 4KB matches the swap size (when the system wants to swap, the units of measure are similar to the size of page table entries). That's part of the reason we got into this mess. The 4KB was selected eons ago. Bummer. There are lots of 4KB sized entries in the TLB, and the CPU design is prefaced on supporting the slow stuff well. There are fewer slots for LargePages. Paul |
#14
|
|||
|
|||
Samsung Magician
On 30/06/2019 01:16, Rene Lamontagne wrote:
There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene I enabled Rapid Mode as soon as I saw it. Didn't look bad. Why not use it? But I haven't any proof as to the SSD works faster or not. It just works fine. Fokke |
#15
|
|||
|
|||
Samsung Magician
Rene Lamontagne wrote:
On 2019-06-30 10:08 a.m., Rene Lamontagne wrote: On 2019-06-30 5:06 a.m., VanguardLH wrote: Rene Lamontagne wrote: There is a feature in this program to run certain Samsung 840, 850 evo SSDs and some other models in what they call "Rapide Mode". Does any one here have any experience with this feature? googleing returns me some good results and a lot of Not sure results. Benchmarking is not possible, it just gives you screwy results that are meaningless, Apperently it uses part of your memory to do this. Just curious to know if it works as I have 2 of these 850 evo drives but only used for game programs, not really a concern. Thanks, Rene I've used it since my Win7 setup when I got the Samsung EVO-840.* Their benchmarks are skewed.* Other benchmarks weren't so impressive, but they were better with Rapid Mode on.* However, you need to running into those edge-use cases where a RAM cache will help.* I don't do huge video editing to make use of such huge write buffers. Rapid Mode is on my To Do list to disable.* I don't remember if I enabled it in my new Win10 build.* If I did, I'll disable it when it get to that item way down in my computer To Do list.* As far as I can tell, it hasn't hurt me; however, it hasn't helped, either, so there's no point in keep using it. http://www.thessdreview.com/software...e-2-1-testing/ (dated December 19, 2014) ** Samsung˙s RAPID white paper states that RAPID works by analyzing ** ´system traffic and leverages spare system resources (DRAM and CPU) to ** deliver read acceleration through intelligent caching of hot data and ** write optimization through tight coordination with the SSD.ˇ It's one of these whiz-bang features that sound great on paper and get promoted, like how an $800 water-filtering vacuum is so much cleaner than a $190 dry vacuum, providing you can past the junky watery mess you have to flush down the toilet while hoping you don't plug it.* When I watch those commercials, first thing that comes to my mind is "Why aren't the dry vacuum users cleaning more often, so they don't have to pay $800 to get their carpets clean?" I have a 3-second delay to show the multiple boot choice on startup.* I use Macrium Reflect and have it added as a boot-time option, so I need to restore than I don't have to go hunting for a bootable CD or USB drive to bring up Reflect.* I won't notice another half-second for boot time.* However, Samsung's claim is that boot time would be reduced, not increased; else, they'd lose a selling point right there. For the application loat-time test, I opened the Before and After images from the article, and toggled between those two tabs in the web browser. I did see some difference, but it was miniscule.* Excel was the only one that benefited from Rapid Mode.* Photoshop was the one with the biggest change (and biggest was still a tiny change); however, the first file open was faster with Rapid Mode but subsequent file opens were slower. No win there for Rapid Mode.* Besides, who keeps loading and unloading data files as fast as they can? There is no read performance boost from Rapid Mode yet it does add another layer of vulnerability (to failure, not from infection).* This is like adding sTunnel to an old version of Outlook that doesn't have TLS support required by the chosen e-mail provider, just making the chain longer where each added link increase the failure potential. A couple tests had Rapid Mode the winner; however, I'm not running my home PC as a file or application server.* Personally, I prefer a more stable platform.* That's why I don't overclock, too.* Availability and reliablity are more important to me than a few edge-cases I may never encounter to effect a performance boost as a rare-time event. Remember those hybrid HDDs that used SSDs for caching?* Yeah, they were faster than HDD-only drives but cost a lot more and I didn't hear any of its users extoling the benefits after a year's real-world experience with them.* System RAM is faster than flash memory, so the same concept was applied to SSDs: speed them up with a RAM cache.* I kept using HDDs until I could afford to jump to SSDs rather than spend money on incremental hardware changes.* As for RAM caching of SSDs, naw, I'll wait until something comes along that gives real-world and consistent performance boost in some new hardware advance.* Memristor's anyone?* If they ever do show up, it'll be another 10 years before they hit a price point tolerable to consumers.* I'll be dead by then. I'm not running a transaction server used by hundreds of customers where even the tiniest fraction of performance increase increases revenue that will reduce ROI and start generating revenue.* I haven't any wow-factor performance boost in my Win7 (8GB) and Win10 (64GB) with Rapid Mode enabled.* No point in throwing more software and hardware at a solution that doesn't effect any real or experienced benfit. Also, whether it is cost effective depends on how much system memory you have.* One a 4GB system, nope, won't help and probably won't work.* On 8GB, I still wouldn't waste the RAM.* With 32 or 64GB then maybe I'd be willing to waste some RAM on caching the SSD, but by then I'd probably be using a RAMdisk to make a super-fast drive that is far faster than is flash memory and doesn't self-destruct over time due to oxide stress or write volume concern. The only reason why I will keep Samsung Magician is to change the overprovisioning of the SSD; however, apparently just enlarging the unallocated space on the drive will do that, too.* If I find that last claim is true, bye bye Samsung Magician (after first using it to disable Rapid Mode). After reading your post VanguardLH I decided to take One more kick at the cat. I also have Macrium Reflect on my start multi-boot menu, Makes life easier, I decided on a real world write test which I do quite often which is Do a Macrium backup of my C:\ drive to the Samsung 850 evo. these are the results with Rapide mode *OFF* Size 31.6 GB, time 2:20 minutes, CPU use 19%, memory use 2.4 GB. These are the results with Rapide Mode *ON* Size 31.6 GB, time 2.05 minutes, CPU use 21%, memory use 6.5 MB. If my math is still functional that's about a 7 % gain which is about 4 seconds, not worth bothering with, So In my case Rapide mode will remain off permanently. Rene In my mind it was about 4 seconds, using a calculator it is actually 9.8 seconds, My old age is showing. :-) Rene Wouldn't that only apply if you are saving the backups to a drive that is affected by Rapid Mode? As I recall, Rapid Mode is only effected on the disk where is the OS partition(s). If so, why are you backing up the C: drive and saving the backups to C: drive? I always found it rude that Magician stays loaded after I'm done with it and exit it. I don't need any of its features constantly running. Rapid Mode is a driver, so it doesn't need a GUI running in the background all the time. Yes, you could manually right-click on its tray icon to unload it, but the point of software is so you don't have to perform such trivialities. It should have an option "Unload after exiting GUI". I use a batch file to load Magician and leave its console window open. After visually unloading Magician, I press a key in the console window to terminate the pause. Having to hit a key isn't much easier than right-clicking on the tray icon to exit; however, I keep forgetting the program is still running in the background, so the console window reminds me to kill the SamsungMagician.exe process by tapping a key in the console window. I started reading the online manual. Lots of gotchas in there. The manual is at: https://www.samsung.com/semiconducto...umer/magician/ |
|
Thread Tools | |
Display Modes | Rate This Thread |
|
|