Search the Community
Showing results for tags '2133mhz'.
Found 2 results
Casowen posted a topic in General Notebook DiscussionsI dont have a notebook that does xmp or overclocking (dell latitude). So I would be stuck with stock settings, but could that be changed from an SPD Flash or something? Heres what im getting. http://www.kingston.com/datasheets/HX321LS11IB2K2_16.pdf - I would love to see these work at 10-11-12-31 at T2. They are supposed to auto-overclock, though I dont know how that works with dell latitudes. I though about doing gskill 2133 for the command rate of T1 over kingstons T2, but stability is a little more important then extra speed. Unless of course anyone can show me otherwise.
Hello. I've been tinkering with this issue for a couple months now, and wondering if anyone can shed some light or help think of something I have not. Purchased new M17x R4 from Dell, roughly 3 months ago. Unit has i7 3740 CPU and AMD 7970m GPU. I Added 240GB Mushkin SSD for boot and slaved the original 750GB drive into the secondary bay for data. Upgraded original 1600MHz RAM to 16GB (4x4GB) of Kingston's HyperX 2133MHz DDR3. The 2133MHz RAM does run - and passes all stress/performance tests with flying colors but is generally "finicky" at POST. Laptop likes to come on for a few seconds and then shut back off until a successful POST happens after a few tires. Once it's worked once, it works fine from then on - until you change something related to hardware configuration/settings. I read the wiki page for modern BIOS POST process and I'm thinking this "working fine once it's worked" behavior could be due to BIOS building a "hardware table" that it can then just read from instead of enumerating and detecting everything from scratch - until something changes. I have verified to my own satisfaction that it's not defective RAM. I purchased some extra sticks and observed same behavior with only new sticks installed. Another note - the 2133MHz RAM will function without this issue if forced into 1866MHz mode in BIOS (but I don't want to do that unless I absolutely have to.) Now, for most purposes, it's fine to just leave everything as-is and once I've tried a few power-ups and gotten a successful POST (sometimes it even works the first time after tinkering) all is well. The one place it's an issue is with manual graphics switching. I have the AMD 7970m card with Enduro (which sucks for certain games unless Enduro is disabled.) I don't want to leave the laptop in exclusive dGPU mode all the time because I do use it on battery at times and the discrete mode cuts battery life nearly 50%... so I switch modes with the Fn+F7 and reboot method when I want to play certain games. This causes a full reset and the failing at POST behavior is observed when the unit tries to power back on (in discrete or switchable mode - it's the ACT of switching that's the issue.) Final note - I use svl7's modified A10 BIOS (have verified same behavior with stock BIOS versions A05 and A10) so I have a lot of available options I can tinker with. Here are the things I've found that INCREASE stability (lessen the chances of the failure at POST, or the number of times it will repeat:) Reduce power limits on the CPU in the Performance menu (this helps but even at lowest stock settings, does not quite alleviate 100%) In advanced CPU power options menu, change boot performance mode from "Max Performance" to "Max Battery." In advanced CPU power options menu, apply a current limit setting of 160 (20 amps) to 2nd plane current limit (integrated GPU.) The fact that these things make an impact suggests that available power at time of POST may be a component of the problem. I do own 2x 240w PSU's and same behavior is observed with both, as is on battery. I wonder if the issue could have to do with the 7970m card not supporting power gating? I have observed that if I hang out in BIOS for a while, the 7970 card gets extremely hot - so it's definitely sucking back the juice at boot time! Once I exit and Windows begins to load, the GPU fan will then spin up fast for a few seconds to cool the chip. It runs cool then-on unless under stress. Adding voltage to the RAM (1.55v vs. the standard 1.5v) DOES NOT help, and in fact decreases stability. If anyone has any thoughts or suggestions, I'd much appreciate! -Adieu, Angie.