Search Results

Search found 4516 results on 181 pages for 'vincent low'.

Page 89/181 | < Previous Page | 85 86 87 88 89 90 91 92 93 94 95 96  | Next Page >

  • NVRAM for journals on Linux?

    - by symcbean
    I've been thinking about ways of speeding up disk I/O, and one of the bottlenecks I keep coming back to is the journal. There's an obvious benefit to using an SSD for the journal - over and above just write caching unless of course I just disable the journal with the write cache (after all devicemapper doesn't seem to support barriers). In order to get the benefits from using a BB write cache on the controller, then I'd need to disable journalling - but then the OS should try to fsck the system after an outage. Of course if the OS knows what's in the batter-backed memory then it could use it as the journal - but that means it must be exposed as a block device and only be under the control of the operating system. However I've not been able to find a suitable low-cost device (no, write-levelling for Flash is not adequate for a journal, at least one which uses Smartmedia). While there's no end of flash devices, disk/array controllers with BB write caches, so far I've not found anything which just gives me non-volatile memory addressable as a block storage device.

    Read the article

  • Web application and remote storage of files

    - by Matt
    Hi have a web application that can store lots and lots of files on the server. i.e. users upload data to it. The files are stored below a particular storage path. The web host will be an IBM xseries 345. However, the disks are really expensive so we would like to put the files onto a less expensive server. Now here is the question. Should I use an NFS mount on the IBM server of a path on the storage server? Or should I write some scripts to upload the files to the storage server instead. Both the storage server and the web host are on the same network. Only the web server is visible to the world. Is NFS performance suitable for an expected low to moderately loaded server?

    Read the article

  • Tools to automate recording streaming radio

    - by Stan
    Is there any tool that can automate recording online streaming radio? I've been use totalrecorder which it has below upside: 1. Handy scheduler. 2. Support create recording templates, so I can customize some high/low quality recording. The downside are it requires to open the streaming radio in browser and can't have another sound source. It's recording what comes out from the speaker. What I am looking for is given a online radio url, and the tool can record the audio stream. No matter if I am playing any other music or not. Thanks.

    Read the article

  • Which is generally considered faster or best practice: symlinks or Apache aliases?

    - by Christopher W. Allen-Poole
    I'm curious as to what most people's views are on this subject. Personally, I will almost always prefer symlinks unless I have no other option -- I find that it is far more obvious when someone is navigating the file system, but, on the other hand aliasing is more platform independent. Windows XP, for example, doesn't have anything remotely comparable to symlinks (NTFS junctions are not interpreted correctly by at least some environments), which means that anything which relies on symlinks in a *nix based system cannot be transferred. (I know that Windows 64x OS's have symlinks, but I've not seen if they can be read correctly by the environments previously mentioned) In addition to this, I was also wondering which is considered faster. Is this even possible to know? Do you have a conjecture? I would imagine that since symlinks are generally more low-level than Apache it would make sense that they would be referenced faster, but, on the other hand, I would guess that Apache is required to do a lookup in either case so it would be disk read dependent.

    Read the article

  • reduce memory footprint of java virtual machine

    - by Lorenzo Boccaccia
    I've a citrix server where multiple users use a multiple java application. Is there a way to reduce the memory footprint of the jvm itself? The max heap is already set fairly low (64MB), as the permgen (32MB) space and we're to the point that the jvm itself uses way more memory than the application itself (the committed area is around 350MB) I'm looking for a way to reduce the jvm ram usage or to make the all the applications run within the same jvm or any other way of sharing common pages between running jvm (if possible) or try switch to switch to a jvm if a jvm exists having optimizations relative to this scenario currently using windows 2003 server and sun java virtual machine 1.6

    Read the article

  • Transferring websites from x64 to x86 server

    - by Ke
    Hi, I run a x64 staging server here along with the following: Solr Java etc. However, I am about to get a linode vps for production and quickly realising that x86 is the way to go for their lowest RAM package (thinking to upgrade later). My staging server is x64 with 12gb ram, so going down to 300mb ram is going to feel devilishly slow ;/ Here are my questions: 1) Will I have problems transferring my scripts, dbs etc from a x64 to x86 server? e.g. solr indexes 2) Is it worth going for the x86 package? I am probably going to upgrade later down the line and x64 might be better for the servers with more RAM? should I stick with x64 instead as there isnt much difference when using with low RAM? Cheers Ke

    Read the article

  • Free DNS software with failover support?

    - by Lin
    I'm looking for DNS software that can accomplish the following: Check health of all A records at set intervals If server is unresponsive after multiple successive checks, replace A record with a working server When a server is down, check it periodically. Once it's up, restore normal A records Here's an equivalent I thought of: Run DNS servers with very low TTL (minutes) Use a cron job to periodically query all webservers Use sed to replace A records if need be, and then restart DNS server I have a hard time believing there isn't already something that can accomplish the above. I'm not looking for a paid service, and I'm restricted to anything I can run with root access to a VPS. Any suggestions would be great. Thanks!

    Read the article

  • Ctrl key is broken on HP Envy is broken, where can I find a replacement

    - by NewProger
    I work as a developer so I have to use key combinations a lot. And I have HP Envy laptop. And the Ctrl key is broken for the second time. First time I just took one from my friend. But I don't have any more friends who are willing to sacrifice their Ctrl key Anyone know where I can find one? (or rather a bunch because they are so weak and low quality) I tried to contact HP support but they did everything to prevent people from doing it. And it is impossible to reach HP support. And in my case where warranty is expired it is not possible at all according to their rules. Also I tried Googling but found nothing

    Read the article

  • Removing RAID on a DELL power-edge server

    - by Simon Callan
    We have a Dell Power-edge T110 server, with either a PERC S100 or S300 raid controller (I suspect S100), and 2 x 500GB SATA hard discs. These discs are running in RAID-1 configuration, (either 1 or 2 virtual discs), and on top of this, we have the C and D drives. As we are running low on disc space, we are looking at dropping the RAID to get some disc space back. Is it possible to split the discs back into 2 seperate hards discs that the OS can see, without losing all the data on the drives? Even better, would it be possible to split the D: drive, leaving the OS C: drive in RAID1 configuration?

    Read the article

  • What lasts longer: Data stored on non-volatile flash RAM, optical media, or magnetic disk?

    - by Chris W. Rea
    What lasts longer: Data stored on non-volatile flash RAM (USB stick or SD cards?), optical media (CD, DVD, or Blu-Ray?), or magnetic disk (floppies, hard drives?) My gut tells me optical media, but I'm not sure. Furthermore, which of those digital media would be most suitable for long-term data storage where environmental issues are unknown, such as low/high temperature or humidity? For example, what digital media could be stored in a basement, attic, or time capsule, and be expected to survive a reasonably long time? e.g. a lifetime, and then some. Update: Looks like optical media and magnetic tape each have one vote below. Does anybody else have an opinion or know of a study comparing the two?

    Read the article

  • faster ( squid + apache httpd + apache tomcat )

    - by letronje
    We have a production setup where we have Squid in the front(caching images, js, css, etc) Apache httpd in the middle(prefork + mod_rewrite + mod_jk/AJP + mod_deflate + mod_php(few php pages)) Apache tomcat 5.5 at the end serving all the dynamic stuff. What would be the best way to reduce the overhead of having 3 servers in the request path ? Wondering if replacing httpd with a faster web server like nginx/lighttpd will help. httpd right now does the job of url rewriting(for clean urls) and talking to tomcat(via mod_jk) and compressing output(mod_deflate) and serving some low traffic php pages. What would be ideal replacement for httpd given that we need these features? Is there a way to replace (squid + apache) with a single entity that does caching well (like squid) for static stuff, rewrites url, compresses response and forwards dynamic stuff directly to tomcat ? heard abt varnish cache, wondering if it can help.

    Read the article

  • small light laptop with docking station that has a PCI(e) slot

    - by laramichaels
    Hi everyone, I am looking for a new laptop/netbook and I have two simple requirements: must be small (10-12in screen) and light (<=3-3.5pounds); must work with a dock into which I can put a PCI or PCI Express card. Basically, I have no performance requirements. Discontinued or low-end are fine. All I want is something easy to carry that when at work I can use a PCI/PCI Express card with to do multi-head. Eg, if an Acer Aspire 1410 fitted the second requirement I would by one in a blink. The only solution with which I am acquainted are Thinkpads that can go into the Thinkpad Advanced Dock. Do other laptop (or netbook brands) also offer the possibility of using Although this is not a "requirement" per se, I would rather hear about options that would not involve buying the latest-greatest-+$2k model... : ) Thank you for any advice! ~lara

    Read the article

  • Why do my download speeds drastically vary during a download?

    - by J. Anthony Carter
    I watch the download speed rise and fall like waves in a storm. At night, during low bandwidth usage I have achieve speeds as high as 3.23 M/sec but the watch them decline to 250 K/sec. and then climb back up. Over and over. During the day my best is around 1.67 M/sec with lows into the 65 K/sec. On top of this, why does a download need to slow down when approaching the end of the download? It's not like a multi-hundred ton train needing to decrease speed as it approaches the station.

    Read the article

  • Perform action based on load avg

    - by sfx
    I'm running some web applications on an debian server and have to struggle with ddos attacks sometimes. It's eating up all my resources and I can't ssh anymore into the server. An idea was to drop all connections if the load avg is too high, so there are still resources for me and accept new connections if the load avg is low enough. Since this has to work under heavy load I'm afraid a cronjob wouldn't be fast enough or take too much resources. tl;dr: Is there a way to configure the behavior if the load avg is above a specific threshold?

    Read the article

  • How do I get Windows 7 wallpaper to display the company logo properly?

    - by David Silva Smith
    Windows 7 is not displaying our company background properly. Curves show pixelation and straight lines are jagged. I'm working with a scalable vector graphics (SVG) image that I've exported to the same resolution (pixel dimensions, to be technical) as the desktop, which is 1440x900. I have tried exporting the image as a .png, .jpg, and .bmp. All of these look correct in an image viewing program, such as Windows Photo Viewer and Paint, but when I set the Windows background to these images, curves show pixelation and straight lines are jagged. Reading online, it seems that behind the scenes, Windows is converting the image to a .jpg with low quality compression, which is causing the issue. I've tried setting the image as a background through Internet Explorer, saving it as a .jpg, and putting the file in the Windows photo directory as suggested in some online forums, but none of those solutions have fixed my issue.

    Read the article

  • LAN speeds and firewall/switch connections

    - by microchasm
    I have a small network with about ten users. All workstations flow into a Dell PowerConnect 3424 which then has a single link to a SonicWALL firewall and from there to a cable modem. More important than internet connectivity is speed between machines (specifically a Windows Server box on the LAN which everyone uses simultaneously). I believe the 3424 has gigabit connections, but they look like they're for stacking. Is there a way to test the speeds on the LAN to see where the speeds are at? Is there any low-hanging fruit insofar as increasing speeds?

    Read the article

  • 912 stream processor available in OpenCL

    - by tugrul büyükisik
    I am thinking of assembling this system: AMD CPU (A8-3870 APU which has Radeon HD 6550D inside: 400 stream processors:xxx GFLOPS) nearly 110$ AMD Graphics card: HD 7750 (512 stream processors:819 GFLOPS peak performance) nearly 170$ Appropriate ram (1600MHz bus) Mainboard What GFLOPS level can I reach as a stable mode with using OpenCL and similar programs? Can I use all 912 stream processors at the same time? I am not trying to do a VS question. I need to know what could be better for scientific computing (%75 of the time) and gaming (%25 of the time) because I have a low budget. With "scientific calculations" I mean fluid dynamics/solid state physics simulating; with games I mean those that need openCL and PhysX.

    Read the article

  • How can I convert videos for Sony Xperia U on my Linux box?

    - by Konstantin
    I would like to send a few videos for my girlfriend, but I would ensure they are playable on her mobile phone. I want to carry out the encoding tasks on my Arch Linux box, because my PC is a low end. I can use only ffmpeg or mencoder or other CLI tool. She can play youtube videos but I don't want to upload such videos to youtube. Are there any presets for specific mobile phones? And when she simply clicks a video direct link will it play directly on the cellphone or should I embed it in a html5 / flash player, like on youtube?

    Read the article

  • subset in geom_point SOMETIMES returns full dataset, instead of none.

    - by Andreas
    I ask the following in the hope that someone might come up with a generic description about the problem.Basically I have no idea whats wrong with my code. When I run the code below, plot nr. 8 turns out wrong. Specifically the subset in geom_point does not work the way it should. (update: With plot nr. 8 the whole dataset is plottet, instead of only the subset). If somebody can tell me what the problem is, I'll update this post. SOdata <- structure(list(id = 10:55, one = c(7L, 8L, 7L, NA, 7L, 8L, 5L, 7L, 7L, 8L, NA, 10L, 8L, NA, NA, NA, NA, 6L, 5L, 6L, 8L, 4L, 7L, 6L, 9L, 7L, 5L, 6L, 7L, 6L, 5L, 8L, 8L, 7L, 7L, 6L, 6L, 8L, 6L, 8L, 8L, 7L, 7L, 5L, 5L, 8L), two = c(7L, NA, 8L, NA, 10L, 10L, 8L, 9L, 4L, 10L, NA, 10L, 9L, NA, NA, NA, NA, 7L, 8L, 9L, 10L, 9L, 8L, 8L, 8L, 8L, 8L, 9L, 10L, 8L, 8L, 8L, 10L, 9L, 10L, 8L, 9L, 10L, 8L, 8L, 7L, 10L, 8L, 9L, 7L, 9L), three = c(7L, 10L, 7L, NA, 10L, 10L, NA, 10L, NA, NA, NA, NA, 10L, NA, NA, 4L, NA, 7L, 7L, 4L, 10L, 10L, 7L, 4L, 7L, NA, 10L, 4L, 7L, 7L, 7L, 10L, 10L, 7L, 10L, 4L, 10L, 10L, 10L, 4L, 10L, 10L, 10L, 10L, 7L, 10L), four = c(7L, 10L, 4L, NA, 10L, 7L, NA, 7L, NA, NA, NA, NA, 10L, NA, NA, 4L, NA, 10L, 10L, 7L, 10L, 10L, 7L, 7L, 7L, NA, 10L, 7L, 4L, 10L, 4L, 7L, 10L, 2L, 10L, 4L, 12L, 4L, 7L, 10L, 10L, 12L, 12L, 4L, 7L, 10L), five = c(7L, NA, 6L, NA, 8L, 8L, 7L, NA, 9L, NA, NA, NA, 9L, NA, NA, NA, NA, 7L, 8L, NA, NA, 7L, 7L, 4L, NA, NA, NA, NA, 5L, 6L, 5L, 7L, 7L, 6L, 9L, NA, 10L, 7L, 8L, 5L, 7L, 10L, 7L, 4L, 5L, 10L), six = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L), .Label = c("2010-05-25", "2010-05-27", "2010-06-07"), class = "factor"), seven = c(0.777777777777778, 0.833333333333333, 0.333333333333333, 0.888888888888889, 0.5, 0.888888888888889, 0.777777777777778, 0.722222222222222, 0.277777777777778, 0.611111111111111, 0.722222222222222, 1, 0.888888888888889, 0.722222222222222, 0.555555555555556, NA, 0, 0.666666666666667, 0.666666666666667, 0.833333333333333, 0.833333333333333, 0.833333333333333, 0.833333333333333, 0.722222222222222, 0.833333333333333, 0.888888888888889, 0.666666666666667, 1, 0.777777777777778, 0.722222222222222, 0.5, 0.833333333333333, 0.722222222222222, 0.388888888888889, 0.722222222222222, 1, 0.611111111111111, 0.777777777777778, 0.722222222222222, 0.944444444444444, 0.555555555555556, 0.666666666666667, 0.722222222222222, 0.444444444444444, 0.333333333333333, 0.777777777777778), eight = c(0.666666666666667, 0.333333333333333, 0.833333333333333, 0.666666666666667, 1, 1, 0.833333333333333, 0.166666666666667, 0.833333333333333, 0.833333333333333, 1, 1, 0.666666666666667, 0.666666666666667, 0.333333333333333, 0.5, 0, 0.666666666666667, 0.5, 1, 0.666666666666667, 0.5, 0.666666666666667, 0.666666666666667, 0.666666666666667, 0.333333333333333, 0.333333333333333, 1, 0.666666666666667, 0.833333333333333, 0.666666666666667, 0.666666666666667, 0.5, 0, 0.833333333333333, 1, 0.666666666666667, 0.5, 0.666666666666667, 0.666666666666667, 0.5, 1, 0.833333333333333, 0.666666666666667, 0.833333333333333, 0.666666666666667), nine = c(0.307692307692308, NA, 0.461538461538462, 0.538461538461538, 1, 0.769230769230769, 0.538461538461538, 0.692307692307692, 0, 0.153846153846154, 0.769230769230769, NA, 0.461538461538462, NA, NA, NA, NA, 0, 0.615384615384615, 0.615384615384615, 0.769230769230769, 0.384615384615385, 0.846153846153846, 0.923076923076923, 0.615384615384615, 0.692307692307692, 0.0769230769230769, 0.846153846153846, 0.384615384615385, 0.384615384615385, 0.461538461538462, 0.384615384615385, 0.461538461538462, NA, 0.923076923076923, 0.692307692307692, 0.615384615384615, 0.615384615384615, 0.769230769230769, 0.0769230769230769, 0.230769230769231, 0.692307692307692, 0.769230769230769, 0.230769230769231, 0.769230769230769, 0.615384615384615), ten = c(0.875, 0.625, 0.375, 0.75, 0.75, 0.75, 0.625, 0.875, 1, 0.125, 1, NA, 0.625, 0.75, 0.75, 0.375, NA, 0.625, 0.5, 0.75, 0.875, 0.625, 0.875, 0.75, 0.625, 0.875, 0.5, 0.75, 0, 0.5, 0.875, 1, 0.75, 0.125, 0.5, 0.5, 0.5, 0.625, 0.375, 0.625, 0.625, 0.75, 0.875, 0.375, 0, 0.875), elleven = c(1, 0.8, 0.7, 0.9, 0, 1, 0.9, 0.5, 0, 0.8, 0.8, NA, 0.8, NA, NA, 0.8, NA, 0.4, 0.8, 0.5, 1, 0.4, 0.5, 0.9, 0.8, 1, 0.8, 0.5, 0.3, 0.9, 0.2, 1, 0.8, 0.1, 1, 0.8, 0.5, 0.2, 0.7, 0.8, 1, 0.9, 0.6, 0.8, 0.2, 1), twelve = c(0.666666666666667, NA, 0.133333333333333, 1, 1, 0.8, 0.4, 0.733333333333333, NA, 0.933333333333333, NA, NA, 0.6, 0.533333333333333, NA, 0.533333333333333, NA, 0, 0.6, 0.533333333333333, 0.733333333333333, 0.6, 0.733333333333333, 0.666666666666667, 0.533333333333333, 0.733333333333333, 0.466666666666667, 0.733333333333333, 1, 0.733333333333333, 0.666666666666667, 0.533333333333333, NA, 0.533333333333333, 0.6, 0.866666666666667, 0.466666666666667, 0.533333333333333, 0.333333333333333, 0.6, 0.6, 0.866666666666667, 0.666666666666667, 0.6, 0.6, 0.533333333333333)), .Names = c("id", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "ten", "elleven", "twelve"), class = "data.frame", row.names = c(NA, -46L)) iqr <- function(x, ...) { qs <- quantile(as.numeric(x), c(0.25, 0.5, 0.75), na.rm = T) names(qs) <- c("ymin", "y", "ymax") qs } magic <- function(y, ...) { high <- median(SOdata[[y]], na.rm=T)+1.5*sd(SOdata[[y]],na.rm=T) low <- median(SOdata[[y]], na.rm=T)-1.5*sd(SOdata[[y]],na.rm=T) ggplot(SOdata, aes_string(x="six", y=y))+ stat_summary(fun.data="iqr", geom="crossbar", fill="grey", alpha=0.3)+ geom_point(data = SOdata[SOdata[[y]] > high,], position=position_jitter(w=0.1, h=0),col="green", alpha=0.5)+ geom_point(data = SOdata[SOdata[[y]] < low,], position=position_jitter(w=0.1, h=0),col="red", alpha=0.5)+ stat_summary(fun.y=median, geom="point",shape=18 ,size=4, col="orange") } for (i in names(SOdata)[-c(1,7)]) { p<- magic(i) ggsave(paste("magig_plot_",i,".png",sep=""), plot=p, height=3.5, width=5.5) }

    Read the article

  • SQL Server cluster performance baseline

    - by Dwight T
    Currently I'm tasked with getting a good performance baseline on a SQL 2005 cluster. The main db on the server is for Sharepoint, but I would like to add other dbs on the cluster. I do have access to Quest's Performance Analysis tool to help. What are key factors to look at to see if the cluster can handle additional dbs? Do you look at different performance indicators for a cluster vs a stand alone sql server? One db will be a low usage transactional db and a read only db that is used for sales data. Thanks Dwight

    Read the article

  • Is there a SIP provider in the UK which provides the P-Asserted-Identity header?

    - by nbolton
    In the US, Flowroute (low cost SIP trunking provider) provides P-Asserted-Identity in the SIP invite request header (example screenshots). It also allows you to set the caller ID for outgoing calls, for example by using the follow in extensions.conf for Asterisk: exten => id,n,Set(CALLERID(all)=123) However, in the UK, I've tried a couple of SIP providers and none of them let me do either of those things (see P-Asserted-Identity or set the caller-ID). Is this because of some sort of restriction in the UK phone networks, or is it only available to really expensive SIP trunking providers?

    Read the article

  • How do I get Chrome to not try to open in Windows 8 mode?

    - by Citrus Rain
    It's not that I'm trying to make it more like 7. That's not the problem. My problem is that whenever I click the Google Chrome icon on my taskbar or even the desktop, Windows brings up this screen-spanning message box saying that my resolution is too low to launch run Google Chrome, and to set my computer to a higher resolution. (I can't.) It then continues to open Chrome correctly after I hit close. This also happens when I click a link in an external program like skype or psi.

    Read the article

  • Is It Possible to Change Default Windows Idle Time for Task Scheduler?

    - by alharaka
    From the official Microsoft docs: Detecting the Idle State The Task Scheduler service will verify that the computer is in an idle state every 15 minutes. The computer is considered idle if all the processors and all the disks were idle for more than 90% of the past 15 minutes and if there is no keyboard or mouse input during this period of time. Besides, any presentation type application that sets the ES_DISPLAY_REQUIRED flag will make Task Scheduler to not consider the system as being idle. In Windows 7, Task Scheduler considers a processor as idle even when low priority threads (thread priority < normal) execute. Is there any way to change the time to less than 15 minutes minutes? Am I assuming this hard-coded and impossible? My Google-fu has eluded me so far and I found nothing, but wanted to check here before giving up.

    Read the article

  • HPC Cluster planning workflow?

    - by Veronica
    After three days of intensive Google searching, I have not found any high-level workflow of how to build a low profile - cheap - computing cluster (we are not interested in HA yet). This is just a front-end plus a node for now. We want to start small with rockscluster, provide a web-based server for offering services, and then add nodes as our budget increases. We're small company, so we haven't enough human resources to implement it smoothly. Here are some facts about our environment: Our hardware is not constant (we will add nodes). Our workload will vary (in the order from 200Mb - 1Tb) Our software will change (scientific applications for data mining) Do you know any visual workflow, worksheet, chart, describing the general necessary steps to begin our cluster planning?

    Read the article

  • DSP - Filtering in the frequency domain via FFT

    - by Trap
    I've been playing around a little with the Exocortex implementation of the FFT, but I'm having some problems. Whenever I modify the amplitudes of the frequency bins before calling the iFFT the resulting signal contains some clicks and pops, especially when low frequencies are present in the signal (like drums or basses). However, this does not happen if I attenuate all the bins by the same factor. Let me put an example of the output buffer of a 4-sample FFT: // Bin 0 (DC) FFTOut[0] = 0.0000610351563 FFTOut[1] = 0.0 // Bin 1 FFTOut[2] = 0.000331878662 FFTOut[3] = 0.000629425049 // Bin 2 FFTOut[4] = -0.0000381469727 FFTOut[5] = 0.0 // Bin 3, this is the first and only negative frequency bin. FFTOut[6] = 0.000331878662 FFTOut[7] = -0.000629425049 The output is composed of pairs of floats, each representing the real and imaginay parts of a single bin. So, bin 0 (array indexes 0, 1) would represent the real and imaginary parts of the DC frequency. As you can see, bins 1 and 3 both have the same values, (except for the sign of the Im part), so I guess bin 3 is the first negative frequency, and finally indexes (4, 5) would be the last positive frequency bin. Then to attenuate the frequency bin 1 this is what I do: // Attenuate the 'positive' bin FFTOut[2] *= 0.5; FFTOut[3] *= 0.5; // Attenuate its corresponding negative bin. FFTOut[6] *= 0.5; FFTOut[7] *= 0.5; For the actual tests I'm using a 1024-length FFT and I always provide all the samples so no 0-padding is needed. // Attenuate var halfSize = fftWindowLength / 2; float leftFreq = 0f; float rightFreq = 22050f; for( var c = 1; c < halfSize; c++ ) { var freq = c * (44100d / halfSize); // Calc. positive and negative frequency indexes. var k = c * 2; var nk = (fftWindowLength - c) * 2; // This kind of attenuation corresponds to a high-pass filter. // The attenuation at the transition band is linearly applied, could // this be the cause of the distortion of low frequencies? var attn = (freq < leftFreq) ? 0 : (freq < rightFreq) ? ((freq - leftFreq) / (rightFreq - leftFreq)) : 1; // Attenuate positive and negative bins. mFFTOut[ k ] *= (float)attn; mFFTOut[ k + 1 ] *= (float)attn; mFFTOut[ nk ] *= (float)attn; mFFTOut[ nk + 1 ] *= (float)attn; } Obviously I'm doing something wrong but can't figure out what. I don't want to use the FFT output as a means to generate a set of FIR coefficients since I'm trying to implement a very basic dynamic equalizer. What's the correct way to filter in the frequency domain? what I'm missing? Thanks in advance.

    Read the article

< Previous Page | 85 86 87 88 89 90 91 92 93 94 95 96  | Next Page >