You hit the nail right on the head Drew! My ISP (Mindspring) has just updated my local POP (Melbourne, Florida) to V.90 from X2. Using my USR X2 upgraded to V.90, I am able to connect at 50666. I have experimented with the following K56Flex modems: Motorola ModemSurfr, Boca Research 56K Internal, Wisecom Accelerator Pro. I have tried using different versions of the same K56Flex to V.90 upgrade protocol that's currently available (listings found on the 56K site: http://www.56k.com/buyer/itu56k.shtml ) **Note: the Upgrade V.90/K56Flex "Wizard" as listed by each site will work with just about any K56Flex modem (be sure to backup your firmware first). The latest version I have seen is from the Genius 56K site (version 2.083, dated 5-12-98). Updating each of the modems firmware, I have NOT been able to connect any of the modems listed above at speeds over 28800. So, I believe that even though there is supposed to be a "standard", it obviously needs some fine tuning before all modems will in fact be equal in the V.90 world. ...Another interesting note here: When I connected to Mindspring before they upgraded to V.90, my connect speed would always read 53300, with the average file download speed being approximately 5.7K to 5.9Kbs. I am now connecting at 50666 and the average download speed (using System Monitor) still shows between 5.7K and 5.9K. Hmmmm, I wonder if USR kinda overestimated the connection speed for their X2 protocol modems! ...One final note here: I originally sign up with K56Flex protocol ISP's in my area using the above 3 modems. I could never get a connection speed higher than 34000. My local Telco (Bell South) said that I had bad wiring. I switched to USR and it's X2 after reading that Boardwatch article and have been crusing at 50K speeds since then! Earl Douglass [log in to unmask] -----Original Message----- From: Drew Dunn <[log in to unmask]> Date: Saturday, May 23, 1998 1:48 AM Subject: Re: [PCBUILD] check 56k upgrade results before buying > >My understanding is that the standard has been released. There does, >however, appear to be issues in its implementation. From a specification >point of view, a standard is a standard, but anyone who's been around for a >few years knows that the implementation doesn't always work out so well. >Thus we find that Rockwell and Lucent chipsets may not like an ISP's USR >hardware, while USR chipset-based modems won't connect correctly to >Livingston hardware. > <snip>